You are on page 1of 422

Power Systems

A.Penin

Analysis of
Electrical Circuits
with Variable Load
Regime Parameters
Projective Geometry Method
Second Edition

Power Systems

More information about this series at http://www.springer.com/series/4622

A. Penin

Analysis of Electrical Circuits


with Variable Load Regime
Parameters
Projective Geometry Method
Second Edition

123

A. Penin
D. Ghitu Institute of Electronic
Engineering and Nanotechnologies
Academy of Sciences of Moldova
Chiinu
Moldova

ISSN 1612-1287
Power Systems
ISBN 978-3-319-28450-7
DOI 10.1007/978-3-319-28451-4

ISSN 1860-4676

(electronic)

ISBN 978-3-319-28451-4

(eBook)

Library of Congress Control Number: 2015935416


Springer International Publishing Switzerland 2015, 2016
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microlms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specic statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by SpringerNature
The registered company is Springer International Publishing AG Switzerland

To my wife Lyudmila for many years


of hope of success

Foreword

The classical electric circuit theory is widely presented in modern educational and
scientic publications. The properties of circuits, which allow simplifying the
analysis of complex networks in the case of change of one (or several) resistance or
load, are known. These properties make appropriate sections in the circuit theory
books. The problems of application and development of basic properties of circuits
in an analytical view draw the attention of researchers at present times, despite
opportunities of simulation computer systems.
It occurs in science that someone looks at traditional problems from another
viewpoint and a whole new direction with a new mathematical apparatus appears in
this traditional area of science and engineering. The book, written by A. Penin, is an
example of such a new approach in circuit theory or methods of their analysis.
This book includes the authors main publications, results of more than 30-year
searches, reflections, and doubts. All this determine a new, unusual approach to the
solution of a number of tasks of circuit theory.
The combination of such concepts as circuit theory and projective geometry (in
general, non-Euclidean geometry), is a new representation in electrical engineering
and circuit theory. There are almost no modern publications of other authors using
such a geometrical approach for circuit analysis. At the same time, similar geometrical methods are widely used in other areas of science. For a wide audience, it
is necessary to explain that geometry is a much deeper concept than a simple
graphical representation of studied dependences.
In the foreword, there is no opportunity and no need to speak about the content
of the authors main results, but one important circumstance should be noted. The
used mathematical apparatus gives a successful interpretation for observed
dependences, validation of used denitions and concepts, removes a tradition, and
their formal introduction. Therefore, the book supplements and develops basic
methods of circuit analysis, makes them evidential, and simplies understanding
of the complicated interconnected processes happening in electric circuits.
The author restricts his analysis by direct current circuits (as models of power
supply systems with one, two, and more loads), which gives the possibility to show

vii

viii

Foreword

in the simplest way the new things or results that geometry can propose for circuit
analysis.
The theorem of the generalized equivalent circuit is such a result. The author
determines the parameters of this equivalent for a network with changeable
parameters and load.
Another result is that mutual changes of resistance and current are set not in the
form of generally used increments (typical for mathematical analysis and a concept
of the derivative) but as well-founded fractionally linear expression. On this basis,
convenient formulas of recalculation of currents turn out. It is possible to carry out
the normalization of running regime parameters and to prove a technique for
comparison of regimes for circuits similar to each other.
Using own scientic and practical experience in power electronics, the author
applies new representations to a number of traditional problems in this area. It is
natural that the author uses idealized models of considered power electronics
appliances and therefore does not give the nished practical decisions. Also, the
possible directions of development of this approach are shown in this book.
As a whole, the book represents a theoretical, methodological, and methodical
interest for those who study circuit theory or work in this area directly. It is possible
to hope that the presented geometrical approach to known tasks and arising new
problems will cause the active permanent interest of experts in various areas of
electrical engineering and radio electronics.
Prof. Anatolie Sidorenko

Preface

The circuit regime analysis is one of the main problems for electric circuit theory.
The nding of the actual (absolute) value of regime parameters (voltage, current,
power, and transformation ratio for different parts of a circuit) is the simplest
analysis task. If a circuit has variable elements (loads and voltage regulators),
additional analysis tasks appear.
The interest in such circuits is dened, in particular, by the state and tendencies
of development of power electronics, modular power supply, or distributed power
supply systems with renewable power sources. Similar devices, in general, represent the complex multiple inputs and multiple outputs systems and their loads can
change from the short circuit to open circuit and further give energy. In turn, the
loads can be subdivided into high priority and additional (ballast) loads. For deniteness, it is possible to accept that such systems, for circuit theory, present linear
mesh circuits of a direct current or multi-port networks.
We will consider some of the arising additional tasks of analysis. For example, it
is important to confront operating regime parameters with characteristic values; that
is, to represent these parameters in the normalized or relative form. In this case, the
informational content of these parameters is increasing; it is possible to appreciate
qualitative characteristics of an operating regime or its effectiveness, to compare
regimes of different circuits, and to set a necessary regime.
The other task of analysis is the determination of the dependence of the regime
parameter changes on the respective change of elements parameters (for example,
the problem of the recalculation of load currents). Thus, it is necessary to set the
form of these changes reasonably; that is, to determine whether these changes are
increments or any other expressions.
Another task of analysis is the denition of the view or character of such an
active circuit with a changeable element (as a power source concerning load); that
is, this circuit shows more property of a voltage source or current source.
In the electric circuit theory, a range of circuits properties, theorems, and
methods is well known, and their use simplies the decision of these problems.

ix

Preface

However, the known approaches do not completely disclose the properties of


such circuits, which reduces the effectiveness of analysis.
The method of analysis for a circuit with variable element parameters is
developed by the author. For interpretation of changes or kinematics of circuit
regimes, projective geometry is used. For example, the known expression has the
typical fractionally linear view for functional dependence of current (or voltage) via
resistance. It gives the grounds for considering this expression as a projective
transformation. The projective transformations preserve an invariant; there is a
cross ratio of four points (a ratio of two proportions) or four values of current and
resistance. The value of this invariant is preserved for all the variables (as a current,
voltage, and resistance) and for parts or sections of a circuit. Thus, this invariant is
accepted as the determination of the regime in the relative form. Therefore, obvious
changes in regime parameters in the form of increments are formal and do not
reflect the substantial aspect of the mutual influences: resistance current.
In general, this geometrical approach grounds the introduction and determination
of required concepts.
The book has an introduction (Chap. 1) and four parts. The disadvantages of
known methods are considered in Chap. 1.
Part I (Chaps. 25) considers electrical circuits with one load. The application
of projective geometry to analysis of an active two-pole is shown in Chap. 2.
The concept of generalized equivalent circuits is introduced in Chap. 3. The
invariant relationships of cascaded two-ports are considered in Chap. 4. In Chap. 5,
the paralleling voltage sources are presented.
Part II (Chaps. 69) considers multi-port circuits. The application of projective
geometry to analysis of an active two-port and three-port is shown in Chap. 6. The
concept of generalized equivalent circuits of multi-port is introduced in Chap. 7.
The recalculation formulas of load currents are obtained in Chap. 8. The invariant
relationships of cascaded four-ports are considered in Chap. 9.
Part III (Chaps. 1012) considers circuits with nonlinear regulation curves. The
voltage regulator regimes are studied in Chap. 10. The load voltage stabilization is
shown in Chap. 11. The pulse-width modulation converters are considered in
Chap. 12.
Part IV (Chaps. 13 and 14) discusses circuits with nonlinear load characteristics.
The concepts of power-source and power-load elements with two-valued characteristics are shown in Chap. 13. Quasi-resonant voltage converters with
self-limitation of current are considered in Chap. 14. The attention to similarity of
characteristics of this converter and some electronic devices is paid.
The book may be useful to those who are interested in the foundations of the
electric circuit theory and also for a professional circle of experts in various areas of
electrical engineering and radio electronics.
The author thanks Prof. A. Scherba (Ukraine) for the attention shown and recommendations to the publication of the rst results. He also thanks Prof. P. Butyrin
(Russia) for long-term constructive criticism and recommendations to the publication of series of papers. He also thanks Prof. V. Mazin (Russia), who shares his

Preface

xi

area of researches, and appreciates scientic editor B. Makarshin (Russia) for a


20-year hard effort on the preparation of paper manuscripts for publication.
The author is grateful to academician D. Ghitu, the founder of Institute of
Electronic Engineering and Nanotechnologies (Moldova), for the given opportunity
to work according to the individual plan of researches. He appreciates Prof.
A. Sidorenko, the director of D. Ghitu Institute of Electronic Engineering and
Nanotechnologies, for permanent interest to the direction of research, support for
this direction and the book publication.
Moldova Republic
October 2015

A. Penin

Contents

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1 Typical Structure and Equivalent Circuits of Power
Supply Systems. Features of Circuits with Variable
Operating Regime Parameters . . . . . . . . . . . . . . . . . . . . .
1.2 Disadvantages of the Well-Known Calculation Methods
of Regime Parameters in the Relative Form for Active
Two-Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1 VoltAmpere Characteristics of an Active
Two-Pole . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2 Regime Parameters in the Relative Form . . . . . . . .
1.2.3 Regime Change in the Relative Form . . . . . . . . . .
1.2.4 Active Two-Port with Changeable Resistance . . . .
1.2.5 Scales of Regime Parameters for Cascaded
Two-Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Analysis of the Traditional Approach to Normalizing of
Regime Parameters for the Voltage Linear Stabilization . . .
1.4 Active Two-Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.1 Volt Characteristics of an Active Two-Port . . . . . .
1.4.2 Traditional Recalculation of the Load Currents. . . .
1.5 Nonlinear Characteristics. . . . . . . . . . . . . . . . . . . . . . . . .
1.5.1 Efciency of Two-Ports with Different Losses . . . .
1.5.2 Characteristic Regimes of Solar Cells . . . . . . . . . .
1.5.3 Quasi-resonant Voltage Converter. . . . . . . . . . . . .
1.5.4 Power-Source and Power-Load Elements. . . . . . . .
1.6 Regulated Voltage Converters . . . . . . . . . . . . . . . . . . . . .
1.6.1 Voltage Regulator with a Limited Capacity Voltage
Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.2 Buck Converter . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.3 Boost Converter . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

...

...

...

.
.
.
.

.
.
.
.

3
4
7
9

...

.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

11
14
14
14
17
17
20
20
21
21

.
.
.
.

.
.
.
.

.
.
.
.

21
23
24
25

xiii

xiv

Contents

Part I
2

Electrical Circuits with one Load. Projective Coordinates


of a Straight Line Point

Operating Regimes of an Active Two-Pole. Display of Projective


Geometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1 VoltAmpere Characteristics of an Active Two-Pole. Afne
and Projective Transformations of Regime Parameters . . . . .
2.1.1 Afne Transformations . . . . . . . . . . . . . . . . . . . . .
2.1.2 Projective Transformations . . . . . . . . . . . . . . . . . .
2.2 VoltAmpere Characteristics of an Active Two-Pole
with a Variable Element . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Thvenin Equivalent Circuit with the Variable
Internal Resistance . . . . . . . . . . . . . . . . . . . . . . . .
2.2.2 Norton Equivalent Circuit with the Variable
Internal Conductivity . . . . . . . . . . . . . . . . . . . . . .
2.3 Regime Symmetry for a Load-Power . . . . . . . . . . . . . . . . .
2.3.1 Symmetry of Consumption and Return of Power . . .
2.3.2 Symmetry Relatively to the Maximum Power Point .
2.3.3 Two Systems of Characteristic Points . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Generalized Equivalent Circuit of an Active Two-Pole
with a Variable Element. . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Circuit with a Series Variable Resistance. . . . . . . . . . . .
3.2.1 Disadvantage of the Known Equivalent Circuit .
3.2.2 Generalized Equivalent Circuit . . . . . . . . . . . . .
3.2.3 Relative Operative Regimes. Recalculation
of the Load Current . . . . . . . . . . . . . . . . . . . .
3.2.4 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 Circuit with a Shunt Variable Conductivity . . . . . . . . . .
3.3.1 Disadvantage of the Known Equivalent Circuit .
3.3.2 Generalized Equivalent Circuit . . . . . . . . . . . . .
3.3.3 Relative Operative Regimes. Recalculation
of the Load Current . . . . . . . . . . . . . . . . . . . .
3.3.4 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4 General Case of an Active Two-Pole with a Variable
Conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.1 Known Equivalent Generator . . . . . . . . . . . . . .
3.4.2 Generalized Equivalent Circuit . . . . . . . . . . . . .
3.4.3 Example of a Circuit. Recalculation of the Load
Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5 Stabilization of the Load Current . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

29

..
..
..

29
29
37

..

42

..

42

.
.
.
.
.
.

.
.
.
.
.
.

44
47
48
50
52
54

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

55
55
56
56
57

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

60
65
70
70
71

.....
.....

74
78

.....
.....
.....

82
83
84

.....
.....
.....

87
91
94

Contents

xv

Two-Port Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Input-Output Conformity of Two-Ports as Afne
Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.1 Conformity of a Two-Port . . . . . . . . . . . . . . . . .
4.1.2 Conformity of Cascaded Two-Ports . . . . . . . . . .
4.2 Input-Output Conformity of Two-Ports as Projective
Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Conformity of a Two-Port . . . . . . . . . . . . . . . . .
4.2.2 Versions of Conformities, Invariants, and Cross
Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3 Conformity of Cascaded Two-Ports . . . . . . . . . .
4.3 Use of Invariant Properties for the Transfer of Measuring
Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Transfer of Signals over an Unstable Two-Port . .
4.3.2 Conductivity Measurement by an Unstable
Two-Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4 Deviation from the Maximum Efciency of a Two-Port . .
4.4.1 Regime Symmetry for the Input Terminals . . . . .
4.4.2 Regime Symmetry for the Output or Load. . . . . .
4.5 Effectiveness of Modular Connections. . . . . . . . . . . . . . .
4.5.1 Complementary Knowledge About a Two-Port . .
4.5.2 Parallel Connection of Two Converters . . . . . . . .
4.5.3 Connection of Two-Ports with the Interaction . . .
4.6 Effectiveness Indices of a Two-Port with Variable Losses .
4.6.1 Problems of Energy Indices . . . . . . . . . . . . . . . .
4.6.2 Inuence of Losses on the Load Power . . . . . . . .
4.6.3 Inuence of Losses on the Efciency . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

....

97

....
....
....

97
97
99

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

119
120
121
123
126
126
127
130
131
131
131
135
138

Paralleling of Limited Capacity Voltage Sources . . . . . . . . .


5.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Initial Relationships . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Inuence of the Load Value on the Current Distribution .
5.3.1 Analysis of Paralleling Voltage Sources . . . . . .
5.3.2 Introduction of Two Concepts . . . . . . . . . . . . .
5.3.3 Comparison of a Loading Regime of Different
Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4 Inuence of the Equalizing Resistance on the Current
Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4.1 Analysis of Paralleling Voltage Sources . . . . . .
5.4.2 Introduction of Two Concepts . . . . . . . . . . . . .
5.4.3 Comparison of a Loading Regime of Different
Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

141
141
141
143
143
145

.
.
.
.
.
.

. . . . 102
. . . . 102
. . . . 106
. . . . 109
. . . . 116
. . . . 116

. . . . . 148
. . . . . 153
. . . . . 153
. . . . . 154
. . . . . 157
. . . . . 162

xvi

Contents

Part II
6

Multi-port Circuits. Projective Coordinates of a Point


on the Plane and Space

Operating Regimes of an Active Multi-port . . . . . . . . . .


6.1 Active Two-Port. Afne and Projective Coordinates
on the Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.1 Afne Coordinates . . . . . . . . . . . . . . . . . .
6.1.2 Particular Case of a Two-Port. Introduction
of the Projective Plane . . . . . . . . . . . . . . .
6.1.3 General Case of a Two-Port. Projective
Coordinates . . . . . . . . . . . . . . . . . . . . . . .
6.2 Projective Coordinates in Space . . . . . . . . . . . . . . .
6.2.1 Particular Case of a Multi-port . . . . . . . . . .
6.2.2 General Case of a Multi-port. The Balanced
Networks . . . . . . . . . . . . . . . . . . . . . . . .
6.3 Projective Coordinates of an Active Two-Port
with Stabilization of Load Voltages . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . 167
. . . . . . . . 167
. . . . . . . . 167
. . . . . . . . 173
. . . . . . . . 175
. . . . . . . . 183
. . . . . . . . 183
. . . . . . . . 190
. . . . . . . . 198
. . . . . . . . 205

Recalculation of Load Currents of Active Multi-ports . . . . .


7.1 Recalculation of Currents for the Case of Load Changes .
7.1.1 Active Two-Port. . . . . . . . . . . . . . . . . . . . . . .
7.1.2 Active Three-Port . . . . . . . . . . . . . . . . . . . . . .
7.2 Recalculation of Currents for the Case of Changes
of Circuit Parameters . . . . . . . . . . . . . . . . . . . . . . . . .
7.2.1 Change of Lateral Conductivity . . . . . . . . . . . .
7.2.2 Change of Longitudinal Conductivity . . . . . . . .
7.3 Comparison of Regimes and Parameters of Active
Two-Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4 Comparison of Regime of Active Two-Ports with Linear
Stabilizations of Load Voltages . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

207
207
207
210

. . . . . 228
. . . . . 235

Passive Multi-port Circuits . . . . . . . . . . . . . . . . . . . . . . .


8.1 Input-Output Conformity of Four-Ports as an Afne
Transformation. . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.2 Input-Output Conformity of Four-Ports as a Projective
Transformation. . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.2.1 Output of a Four-Port . . . . . . . . . . . . . . . . .
8.2.2 Input of a Four-Port . . . . . . . . . . . . . . . . . .
8.2.3 Recalculation of Currents at Load Changes . .
8.2.4 Two Cascaded Four-Port Networks. . . . . . . .
8.2.5 Examples of Calculation . . . . . . . . . . . . . . .
8.3 Transmission of Two Signals Over Three-Wire Line. .
8.3.1 Transmission by Using of Cross-Ratio . . . . .
8.3.2 Transmission by Using of Afne Ratio . . . . .

.
.
.
.
.
.
.
.
.

. . . . . 213
. . . . . 213
. . . . . 219
. . . . . 224

. . . . . . . 237
. . . . . . . 237
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

244
244
246
251
252
254
260
260
262

Contents

xvii

8.4 Input-Output Conformity of a Balanced Six-Port . . . . . . . . . . . 263


References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
9

Generalized Equivalent Circuit of a Multi-port . . . .


9.1 Generalized Equivalent of an Active Two-Port. .
9.1.1 Disadvantages of Known Equivalent. . .
9.1.2 Introduction of the Formal Variant
of a Generalized Equivalent . . . . . . . . .
9.1.3 Introduction of the Principal Variant
of a Generalized Equivalent Circuit . . .
9.2 Generalized Equivalent of an Active Three-Port .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Part III

. . . . . . . . . . . 275
. . . . . . . . . . . 275
. . . . . . . . . . . 275
. . . . . . . . . . . 276
. . . . . . . . . . . 279
. . . . . . . . . . . 282
. . . . . . . . . . . 286

Circuits with Non-Linear Regulation Curves

10 Regulation of Load Voltages . . . . . . . . . . . . . . . .


10.1 Base Model. Display of Conformal Geometry
10.2 Using of Hyperbolic Geometry Model . . . . .
10.2.1 Case of One Load . . . . . . . . . . . . .
10.2.2 Case of Two Loads . . . . . . . . . . . .
10.3 Example . . . . . . . . . . . . . . . . . . . . . . . . . .
10.3.1 Case of One Load . . . . . . . . . . . . .
10.3.2 Case of Two Loads . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

289
289
295
296
299
305
306
308
312

11 Stabilization of Load Voltages . . . . . . . . . . . . . . . . . . . . .


11.1 Analysis of Load Voltage Stabilization Regimes . . . . .
11.1.1 Case of One Load . . . . . . . . . . . . . . . . . . . .
11.1.2 Use of Hyperbolic Geometry . . . . . . . . . . . . .
11.1.3 Case of Two Loads . . . . . . . . . . . . . . . . . . .
11.2 Given Voltage for the First Variable Load and Voltage
Regulation of the Second Given Load. . . . . . . . . . . . .
11.2.1 Use of Hyperbolic Geometry . . . . . . . . . . . . .
11.2.2 Regime Change for the First Given
Load Resistance . . . . . . . . . . . . . . . . . . . . . .
11.2.3 Example . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

313
313
313
315
322

. . . . . . 324
. . . . . . 328
. . . . . . 330
. . . . . . 333
. . . . . . 334

12 Pulse-Width Modulation Regulators . . . . . . . . . . . . . . . . . .


12.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12.2 Regulation Characteristic of Boost Converter . . . . . . . . .
12.3 Regulation Characteristic of BuckBoost Converter . . . .
12.3.1 BuckBoost Converter with an Idealized Choke .
12.3.2 BuckBoost Converter with Losses of Choke . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

337
337
337
348
348
354
357

xviii

Part IV

Contents

Circuits with Non-Linear Load Characteristics

13 Power-Source and Power-Load Elements . . . . . . . . . . . . . . . .


13.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13.2 Two-Valued Regime of a Regulated Converter.
The Concept of a Power-Source and Power-Load Element .
13.3 Inuence of Voltage Source Parameters and Power-Load
Element onto a Power Supply Regime . . . . . . . . . . . . . . .
13.3.1 Ideal Voltage Source . . . . . . . . . . . . . . . . . . . . .
13.3.2 Voltage of a Power Supply with Limited Capacity .
13.3.3 Internal Resistance of a Voltage Source . . . . . . . .
13.3.4 Power of a Power-Load Element . . . . . . . . . . . . .
13.4 Power-Load Element with Losses. . . . . . . . . . . . . . . . . . .
13.4.1 Series Loss Resistance . . . . . . . . . . . . . . . . . . . .
13.4.2 Two-Port Loss Circuit. . . . . . . . . . . . . . . . . . . . .
13.5 Power Supply Line with Losses . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . 361
. . . 361
. . . 361
.
.
.
.
.
.
.
.
.
.

14 Quasi-resonant Voltage Converter with Self-limitation


of Load Current. Similarity of Load Characteristics
of Some Electronic Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.1 Load Curve of an Active Two-Pole with Self-limitation
of the Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.2 Equivalent Generator of an Active Two-Pole
with Self-limitation of Current . . . . . . . . . . . . . . . . . . . . . .
14.3 Deviation from the Maximum Load Power Point . . . . . . . . .
14.4 Symmetrical Load Characteristic for the Full Area of the Load
Voltage Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.5 Asymmetrical Load Characteristics . . . . . . . . . . . . . . . . . . .
14.6 Linearly Hyperbolic Approximation of a Solar Cell
Characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14.6.1 Approximation Problem . . . . . . . . . . . . . . . . . . . .
14.6.2 Formal Linearly Hyperbolic Approximation . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.

363
363
364
370
372
374
374
375
381
388

. . 389
. . 389
. . 392
. . 396
. . 400
. . 401
.
.
.
.

.
.
.
.

405
405
405
410

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415

About the Author

A. Penin has engineering interests related to power electronics. He worked


(19802006) on the design ofce of solid-state electronics of Academy of Sciences
of Moldova; elaboration of power supply systems. From 2006 to the present, he
works in D. Ghitu Institute of Electronic Engineering and Nanotechnologies of
Academy of Sciences of Moldova; he continues the early begun independent
theoretical researches in the electric circuit theory with variable regimes. He is a
senior research assistant, Ph.D. (2011). The author has to his credit more than
60 publications, 40 patents of Moldova, and one European patent.

xix

About the Book

This book introduces electric circuits with variable loads and voltage regulators.
It allows to dene invariant relationships for various parameters of regime and
circuit sections and to prove the concepts characterizing these circuits. Generalized
equivalent circuits are introduced. Projective geometry is used for the interpretation
of changes of operating regime parameters. Expressions of normalized regime
parameters and their changes are presented. Convenient formulas for the calculation
of currents are given. Parallel voltage sources and the cascade connection of
multi-port networks are described. The two-value voltage regulation characteristics
of loads with limited power of voltage source is considered. The book presents the
fundamentals of electric circuits and develops circuit theorems.
The second edition was extended and contains additional chapters to circuits
with nonlinear regulation curves, circuits with nonlinear load characteristics, concepts of power-source and power-load elements with two-valued characteristics,
quasi-resonant voltage converters with self-limitation of current as well as the
similarity of characteristics of converters and electronic devices.
This book is useful to engineers, researchers, and graduate students who are
interested in the basic electric circuit theory and the regulation and monitoring of
power supply systems.

xxi

Chapter 1

Introduction

1.1

Typical Structure and Equivalent Circuits of Power


Supply Systems. Features of Circuits with Variable
Operating Regime Parameters

The circuit regime analysis is one of the main problems for the electric circuit
theory [1, 7]. The nding of the actual (absolute) value of regime parameters
(voltage, current, voltage transformation ratio, and so on), is the simplest analysis
task. In turn, the determination of the maximum load power and maximum efciency is an essential energy problem [6, 14], including distributed power supply
systems. Such systems, shown in Fig. 1.1, contain multiple power sources, energy
storages, direct current DC/DC voltage converters, charge/discharge units,
point-of-load POL regulators, and loads [4, 13].
Usually, the static characteristics of all the incoming units are used for energy
analysis [9, 12] and analog computation of power systems [11]. Therefore, this power
supply system is the complex DC circuit with a given number of input and output
terminals. For example, such a circuit is shown in Fig. 1.2. The voltage DC/DC
converters are simply DC transformers. The resistive network determines losses of all
the converters and supply lines.
Let us consider the feature of the show circuit. The interaction both between
power sources and loads is the main feature of this system. The regime of the loads
and energy storages may change from the energy consumption to return of energy.
The loads may be also subdivided into basic (priority) and additional (buffer) loads.
Therefore, additional analysis tasks appear for these circuits with variable elements (loads, voltage regulators). For example, it is important to confront operating
regime parameters with the characteristic values; that is, to represent these
parameters in the normalized or relative form. In this case, the informational content
of these parameters is increasing; it is possible to appreciate the qualitative characteristics of an operating regime or its effectiveness, to compare regimes of different circuits, to set a necessary regime in the sense of similarity theory.
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_1

Introduction

Fig. 1.1 Typical distributed power supply system

Fig. 1.2 Simplied


equivalent circuit of
distributed power supply
system

Usually, relative expressions are constituted by using of the characteristic values (as
scales) for the corresponding regime parameter [19]. Similarly, the regime change is
dened by the difference or ratio of subsequent and initial regime parameters; there are
changes in the form of times or percents. For example, the open-circuit OC
voltage and short circuit SC current will be the corresponding scales for the load

1.1 Typical Structure and Equivalent Circuits

voltage and current of the simplest circuit. In the same way, the maximum load power
or maximum source power will be the scales for the running load power.
The other task of analysis is the determination of the regime parameter changes
via respective changes of element parameters (for example, the problem of the
recalculation of load currents). Thus, it is necessary to set the form of these changes
reasonably; that is, if these changes are increments or the other expressions.
Also, we have the next task. The change of the load parameters or parameters of
circuit denes the corresponding regime change and its effectiveness. Therefore, a
deeper analysis and introduction of changes in the valid form of both regime
parameters and effectiveness indicators are necessary.
In the electric circuit theory a range of properties, theorems, and methods are
well-known, and their use can simplify the solution of these problems.
However, the known approaches do not completely disclose the properties of the
circuits with variable elements that reduce the efciency of analysis.
Using the equivalent circuit in Fig. 1.2, we will choose the simplest and
important circuits. The analysis of such simple networks shows the disadvantages
of the known methods.

1.2

1.2.1

Disadvantages of the Well-Known Calculation


Methods of Regime Parameters in the Relative Form
for Active Two-Poles
VoltAmpere Characteristics of an Active Two-Pole

The simplest circuit of an active two-pole is shown in Fig. 1.3a. At change of a load
resistance RL from the short circuit SC to open circuit OC, a load straight line or
voltampere characteristic in Fig. 1.3b is given by a linear expression
IL

V0 VL
VL

ILSC  ;
Ri Ri
Ri

where Ri is an internal resistance and ILSC is the SC current.


Fig. 1.3 a Simplest circuit.
b Load straight line

(a)

(b)

1:1

(a)

Introduction

(b)

Fig. 1.4 a Influence of loads to each other. b Influence of voltages sources to each other

In turn, an internal resistance Ri in Fig. 1.4a determines the influence of the load
resistances RL1 ; RL2 to each other. Similarly, the influence between the paralleling
voltage sources V01 ; V02 takes place in Fig. 1.4b.
Let us consider the straight lines of the initial circuit and a similar circuit with the
~0 , R
~ i in Fig. 1.5. The regimes of these circuits will be similar or
other values V
equivalent if the correspondence of the characteristic and running regime parameters takes place. For the given case, this conformity is specied by arrows.

1.2.2

Regime Parameters in the Relative Form

Let us constitute the relative expression of the load straight line for our simplest
circuit in Fig. 1.3. For that, it is possible to use the values ILSC ; V0 as the scales.
We may rewrite Eq. (1.1) in the form
IL
VL
1 :
V0
ILSC

Fig. 1.5 Straight lines of


comparable circuits

1:2

IL

IL

SC

IL

SC

IL

RL
RL

V0

V0

VL, VL

1.2 Disadvantages of the Well-Known Calculation Methods of Regime

Similarly, for the other circuit


~IL
~L
V
1 :
~ILSC
~0
V
From this, the equality of the relative or normalized values for currents and
voltages determines the similarity of regimes
IL

~IL
IL
SC ;
SC
~
IL
IL

VL

~L
VL V
:
~0
V0 V

1:3

Then, Eq. (1.2) obtains the relative view


IL 1  V L:

1:4

Therefore, the relative values I L ; V L allow evaluating the use of the voltage
source for current and voltage of the running regime.
From (1.3), the recalculation formulas of the actual regime parameters are as
follows:
mI

IL ILM

;
~IL ~ILM

mV

VL V0
;
~L V
~0
V

1:5

where mI ; mV are the scales.


If mI mV , a geometrical similarity or, more precisely, the Euclidean equivalence in sense of Euclidean geometry is obtained.
If mI 6 mV , an afne similarity in sense of afne geometry is fullled.
The load-power dependence
PL V L I L V L 1  V L

1:6

determines a parabola in Fig. 1.6.


In turn, the power transfer ratio KP or efciency g of the running regime
KP g

PL
VL
P0

1:7

is a linear function shown in Fig. 1.6.


Let us consider the analogous relationships for the load resistance.
From (1.1), we get the corresponding expression
VL V0

RL
:
Ri RL

1:8

Fig. 1.6 Load power and


efciency via the load voltage

Introduction

1.00

0.75

0.50

PL

0.25

0.00
0.0

0.5

1.0

VL

It is possible to express the relative load resistance by the internal resistance


RL

~L
RL R
:
~i
Ri
R

1:9

Then, we have the relative expression


VL

RL
:
1 RL

1:10

From (1.9), the recalculation formula is as follows:


mR

RL Ri
;
~L R
~i
R

1:11

where mR is the scale.


Hence, we have obtained relative values (1.3), (1.9) and relative expressions
(1.4), (1.10), which coincide for different circuits. It shows as though we do not
have the problem. But, it is possible to remark that relative regime parameters (1.3),
(1.9) have the different quantities. For example, the maximum load power regime
corresponds to the values RL 1; V LM 0:5: That is, inconvenient for the more
complex circuits with a large number of parameters. If the internal resistance
Ri 0, the circuit has not scales for the load current and resistance.
The examples of circuits, which have two characteristic values of regime
parameters, will be shown later. Then, the problem of scales and relative expressions arises.

1.2 Disadvantages of the Well-Known Calculation Methods of Regime

1.2.3

Regime Change in the Relative Form

Let an initial (the rst) regime correspond to values R1L ; VL1 ; IL1 . In turn, a subsequent (the second) regime corresponds to R2L ; VL2 ; IL2 . For convenience, we consider
V0 10; Ri 1. Using (1.1), (1.8), we get the concrete or absolute values of all the
regime parameters shown in Fig. 1.7.
Let us obtain these changes of regime parameters in the absolute and relative
form. In the given case, we use the difference of values that corresponds to the
known variation theorem of resistance and current into a circuit.
Then, we get the relative changes for the currents and voltages
IL2
IL1
DI 21 1
21
I L 0:1:
 SC
SCL
SC
10
IL
IL
IL
VL2 VL1 DVL21
1
21
V L 0:1:


10
V0 V0
V0

1:12

Using (1.9), we constitute the following relative changes for the resistances:
R2L R1L DR21
0:5
L
0:5:


1
Ri Ri
Ri

1:13

In turn, corresponding to the known variation theorem of resistance and current


[2], we have the dependence DIL21 DR21
L in the form

Fig. 1.7 Correspondence of


the regime parameters at
change of the load resistance

RL

RL

1 R21 1.5
L

RL
32

RL

RL

5.666

RL

31

RL

10

IL

IL

21

IL

IL
32

IL

IL

2.5 1.5

IL

31

IL

VL

VL
0

21

VL

VL
7.5

VL
6.5

10 V
L

DIL21 

1
DR21
0:5=2
L =RIN
IL1 
 5 1;
21
1
1 0:5=2
1 DRL =RIN

Introduction

1:14

where R1IN Ri R1L 2 is the input resistance of our circuit for the subsequent
load.
It is possible to rewrite expression (1.14) in the relative view
1
DIL21
DR21
L =RIN

:
1
21
IL
1 DRL =R1IN

Therefore, we get the other determination of the relative changes


DR21
L
0:25;
R1IN

DIL21
0:2:
IL1

1:15

Hence, the practical question arises, which expressions may we use? Because
both initial expressions (1.12), (1.13), and complementary expression (1.15) have a
clear physical sense.
Group Properties of Regime Changes
Let the next regime correspond to the values R3L ; VL3 ; IL3 shown in Fig. 1.7. Then,
corresponding to (1.15), we get the changes relatively to the rst regime
DR31
L
1;
R1IN

DIL31
0:5:
IL1

Let us obtain the changes relatively to the second regime. To do this, we must
calculate R2IN Ri R2L 2:5. Then,
DR32
L
0:6;
R2IN

DIL32
0:375:
IL2

We note that the resultant regime change DIL31 =IL1 0:5 is not expressed
through the initial change DIL21 =IL1 0:2 and intermediate change DIL32 =IL2
0:375 by the group operation (addition or multiplication). In this sense, regime
changes (1.12), (1.13) hold the group properties that are convenient for analysis.
Anyway, the next regime changes will lead to the increase of a number of the
relative values for various parameters and determinations of changes that complicates analysis.

1.2 Disadvantages of the Well-Known Calculation Methods of Regime

Equal Regime Changes for Different Initial Regimes


Let the initial regime correspond to R1L ; VL1 ; IL1 , and subsequent regime corresponds
to R2L ; VL2 ; IL2 . Then, we have some regime changes. Similarly, let the same regime
change correspond to the other initial R3L ; VL3 ; IL3 and subsequent R4L ; VL4 ; IL4 values.
We obtain even more the relative values for various parameters and determinations
of changes.

1.2.4

Active Two-Port with Changeable Resistance

Let an active two-pole A with a load resistance RL1 contain a changeable resistance
RL2 shown in Fig. 1.8a.
Therefore, values R1L2 ; R2L2 and so on determine a family of load straight lines
with the various values of SC currents and OC voltages in Fig. 1.8b. Then, we have
the set of the base or characteristic values as the parameters of the Thvenin
equivalent circuit. Also, the corresponding variations of scales (1.3) complicate
analysis.
Hence, it is necessary to obtain such base values which do not depend on the
variation of an additional load. Also, we may get the relative values and relative
changes of this additional load for comparing regimes for different circuits.

1.2.5

Scales of Regime Parameters for Cascaded Two-Ports

Let us consider a cascade connection of two two-ports TP1; TP2 with a variable
load conductivity YL2 in Fig. 1.9a. At change of load conductivity, we get a family
of load straight lines for various sections; that is, we have some correspondence
between these load lines by dashdot lines shown in Fig. 1.9b.
Fig. 1.8 a Active two-pole
with a changeable resistance
RL2 . b Family of load straight
lines

(a)

(b)

RL2

RL2

I1
SC1

I1

I1
A

V1

RL2

SC2

I1

RL1

RL1

OC2

V1

OC1

V1

V1

10

Fig. 1.9 a Cascade


connection of two two-ports.
b Correspondence of load
straight lines

Introduction

(a)
TP1

V0, I0

TP2

V1, I1

V2, I2

y10
y0

y1

YIN1

YL1=YIN2

I2, I1, I0

(b)

YL2

I1(V1)

I0(V0)
SC

I0

SC

I2(V2)

I1

SC

I0

SC

I2

V1
1

I2

OC

I0

I1

YL2
2

I2

YL2

OC

I1

OC

V1
OC

V2

V0
V2, V1, V0

Our cascaded circuit, relatively to the load YL2 , is similar to the simplest circuit
in Fig. 1.3. Then, the characteristic values V2OC ; I2SC are the scales or base values.
Therefore, the relative values of the load current and voltage are determined by
(1.3)
I2

I2
;
I2SC

V2

V2
:
V2OC

We must remark that the other characteristic values V2SC 0; I2OC 0.


In turn, the internal resistance or conductivity Ri2 1=Yi2 V2OC =I2SC .
Therefore, the relative load is determined by (1.9)
RL2

RL2
:
Ri2

So, we have the following base values for the load section V2 ; I2
V2OC ; I2SC ; V2SC 0; I2OC 0:

1:16

1.2 Disadvantages of the Well-Known Calculation Methods of Regime

11

These values determine the corresponding base values for the other section of
the cascaded circuit. For the section V1 ; I1 , we get
V1OC ; I1SC ; V1SC ; I1OC :

1:17

In turn, for the section V0 ; I0 , the base values are


I0SC ; I0OC :

1:18

The two nonzero base values turn out for all the regime parameters.
Then, there is a problem, how reasonably we may express these parameters for
1
2
, subsequent YL2
and regime changes in the relative form.
the initial regime YL2
It is possible to note that regime changes, as the length of segments on all the
load straight lines have different lengths for usually used Euclidean geometry.

1.3

Analysis of the Traditional Approach to Normalizing


of Regime Parameters for the Voltage Linear
Stabilization

For the illustration of the assigned task, we consider two simple active two-poles
with a load conductivity YL1 in Fig. 1.10. The equation of the load straight line of
the rst active two-pole in Fig. 1.10a is given by
I1 V0  V1 y01 y01 V0  y01 V1 ;
where the conductivity y01 corresponds to the internal resistance of the voltage
source V0 and SC current I1SC y01 V0 .
Then, we use normalized expression (1.2) which contains the two normalized
values
I1 V1
;
:
I1SC V0

(a)
V0

y01

(b)

I1
V1
YL1

V0

V1N

y01

I1

y1N

V1
YL1

Fig. 1.10 a Active two-pole without a voltage stabilization. b Stabilization of a load voltage

12

Introduction

The regimes of two similar circuits with running values of currents I1 ; ~I1 and
~1 will be equivalent or equal to each other if the normalized values of
voltages V1 ; V
currents and voltages (1.3) take place; that is,
~I1
I1
SC ;
SC
~I
I1
1

~1
V1 V
:
~0
V0 V

The equation of the load straight line of the second active two-pole in Fig. 1.10b
is given by
I1

y0N y1N
V0  V1 ;
y0N y1N

where conductivity y1N corresponds to the conductivity of voltage regulator.


It is possible to carry out the normalization by the SC current I1SC if there is an
access to this source at an experimental investigation. Then,


I1
y1N =y0N
V1

1

:
V0
I1SC 1 y1N =y0N

1:19

This expression contains the three normalized values. Therefore, a possible


condition of equal regimes corresponds to the equalities
~I1
I1
SC ;
SC
~I1
I1

~1
V1 V
;
~0
V0 V

y1N ~y1N

:
y0N ~y0N

1:20

If regimes differ, how may we express this difference in the convenient view? It
is not clear, how may we work with the set of these six different values. It would be
convenient to work with one value which characterizes this difference.
If the access to the voltage source is missing, what then must we choose as the
normalizing value? It is possible to normalize by the maximum load current
I1M y0N V0  V1 , when the linear regulator is completely closed. Then, we have
I1
y1N =y0N

:
I1M 1 y1N =y0N

1:21

Therefore, the possible condition of the equal regimes corresponds to the


equalities
~I1
I1

;
I1M ~I1M

y1N ~y1N

:
y0N ~y0N

We again obtain the two normalized values. But there is a contradiction with
condition (1.20) for currents.

1.3 Analysis of the Traditional Approach

13

Fig. 1.11 Equivalent circuit


of the active two-pole with the
voltage stabilization

y01
N

V0

y1N

V1

V1N

In Eq. (1.21) we pass from the value y1N to the voltage V1N of the linear
regulator. Then, we obtain the normalized expression of the regulator
I1
V1N
1
:
I1M
V0  V1
The equivalent circuit in Fig. 1.11 corresponds to this expression.
Even for such a simple circuit there is an uncertainty, how correctly or reasonably we may represent the normalized expression of regime?
The problem becomes complicated even more for a case of two and more loads
with the voltage stabilization. For this, we consider Fig. 1.12.
In the case of two loads (without conductivity yN ), the two equations turns
out to be
y0N y1N
V0  V1 y0N  I2 ;
y1N
y0N y2N
V0  V2 y0N  I1 :
I2
y2N
I1

It is possible to carry out the normalization by the SC current of the voltage


source. These expressions contain six normalized values. If regimes differ, we have
the set of 12 different values.

Fig. 1.12 Active circuit with


the voltage stabilization of
two loads

y1N
y0N

V1

y2N

V0
yN

V2
YL2

YL1

14

Introduction

On the other hand, the normalization by the maximum load currents I1M
y0N V0  V1 ; I2M y0N V0  V2 leads to the reciprocal components I2 =I1M ; I1 =I2M .
These components raise a number of normalized values.
Therefore, the shown examples of the formal normalization do not allow comparing the regimes of different systems.

1.4
1.4.1

Active Two-Port
Volt Characteristics of an Active Two-Port

Let us consider the features of an active two-port with two load conductivities
YL1 ; YL2 in Fig. 1.13a. The interaction between load voltages is observed at change
of loads. Therefore, the variable slope of the load volt characteristics or load straight
lines V2 V1 ; YL1 ; V2 V1 ; YL2 takes place in Fig. 1.13b; the values YL1 ; YL2 are
parameters.
Therefore, such base values as the load SC currents and OC voltages are
changed. In turn, the corresponding variations of scales (1.3) complicate analysis.
Hence, similar to Sect. 1.2.4, it is necessary to obtain such base values that do
not depend on variable loads.

1.4.2

Traditional Recalculation of the Load Currents

Let us lead the base statements of the traditional approach to the recalculation of the
load currents [3]. Expression for the currents I1 ; I2 of Fig. 1.14a has the view
"

I1
I2

Fig. 1.13 a Active two-ports.


b Family of load volt
characteristics

"

Y11 Y12
Y12  Y22

# "

# "
#
I1SC
V1

SC :
V2
I2

(a)

(b)

1:22

V2
YL1

V1
A

YL1

YL2

YL1
1

V2

YL2

YL2

0V

V1

1.4 Active Two-Port

15

Fig. 1.14 a Initial values of


loads. b Increments of loads
concerning to the initial
values

(a)

I2

I1
1

V1

V2

RL1

(b)

RL2

I1

21

RL1

I2

G
Y

RL1

RL2

21

RL2

SC currents I1SC ; I2SC are considered as the initial values I10 ; I20 of the load currents for
the zero load resistances R0L1 0; R0L2 0.
Now, we introduce the rst load resistances R1L1 ; R1L2 as the rst increments
10
DR10
L1 ; DRL2 concerning the initial zero values; that is,
0
10
R1L1 DR10
L1 RL1 DRL1 ;

0
10
R1L2 DR10
L2 RL2 DRL2 :

Therefore, the relationships take place for these subsequent or the rst values of
currents and voltages
1
V11 DR10
L1 I1 ;

1
V22 DR10
L2 I2 :

We rewrite expression (1.22) as




1
10 1
0
I11 Y11 DR10
L1 I1 Y12 DRL2 I2 I1
1
10 1
10 1
0
I2 Y12 DRL1 I1  Y22 DRL2 I2 I2 :

Then, we get


1
10 1
0
1 Y11 DR10
L1 I1  Y12 DRL2 I2 I1
10 1
10 1
Y12 DRL1 I1 1 Y22 DRL2 I2 I20 :

1:23

The solution of Eq. (1.23) gives the subsequent currents


0
0
I10 DR10
L2 Y22 I1 Y12 I2
;
D10
0
10
0
0
I DRL1 Y11 I2 Y12 I1
I21 2
;
D10

I11

1:24

16

Introduction

where the determinant is


2
10
10
10
D10 1 Y11 DR10
L1 1 Y22 DRL2  Y12 DRL1 DRL2 :

Let us carry out the analysis of these relationships. Let the second load resis20
tances R2L1 ; R2L2 be given as the second increments DR20
L1 ; DRL2 concerning the
initial zero values; that is,
0
20
R2L1 DR20
L1 RL1 DRL1 ;

0
20
R2L2 DR20
L2 RL2 DRL2 :

Then,
0
0
I10 DR20
L2 Y22 I1 Y12 I2
;
20
D
0
0
I 0 DR20
L1 Y11 I2 Y12 I1
I22 2
;
D20

I12

where the determinant is


2
20
20
20
D20 1 Y11 DR20
L1 1 Y22 DRL2  Y12 DRL1 DRL2 :

1:25

21
We try to introduce the intermediate changes DR21
L1 ; DRL2 thus
1
21
10
20
R2L1 DR21
L1 RL1 DRL1 DRL1 DRL1 ;
1
21
10
20
R2L2 DR21
L2 RL2 DRL2 DRL2 DRL2 :

Therefore, the denominator D20 will contain both intermediate changes DR21
L1 ;
1
1
DR21
,
and
the
rst
values
of
resistances
R
;
R
.
Thus,
the
structure
of
denomiL2
L1
L2
nator (1.25) shows that group properties are not carried out for intermediate changes
of load resistances. Therefore, subsequent changes must be counted concerning
21
initial zero values. However, if we want to use the changes DR21
L1 ; DRL2 concerning
the values R1L1 ; R1L2 , the two-port with conductivity parameters G is obtained in
Fig. 1.14b. Thus, the recalculation of parameters of a new two-port is required.
Let us introduce variations of currents as increments DI110 ; DI210 concerning the
initial values; that is,
I11 DI110 I10 ;

I21 DI210 I20 :

From (1.23), we get


(

10
10
10
10 0
10 0
1 Y11 DR10
L1 DI1 Y12 DRL2 DI2 Y11 DRL1 I1 Y12 DRL2 I2
10
10
10
10 0
10 0
Y12 DR10
L1 DI1  1 Y22 DRL2 DI2 Y12 DRL1 I1  Y22 DRL2 I2 :

1.4 Active Two-Port

17

The solution of this system gives the increments of currents via the increments of
load resistances
2
10
10
Y11 DR10
Y12 DR10
L1 Y11 Y22  Y12 DRL1 DRL2 0
L2 0
I

I ;
1
D10
D10 2
2
10
10
Y22 DR10
Y12 DR10
L2 Y11 Y22  Y12 DRL1 DRL2 0
L1 0

I

I ;
2
D10
D10 1

DI110
DI210

1:26

Let us compare these expressions with expression (1.14). We cannot already


directly introduce the relative changes of type (1.15).
It is possible to draw the conclusion that it is necessary to carry out a deeper
research of such circuits to obtain the relative expressions of regimes and its
changes.

1.5

Nonlinear Characteristics

Besides the considered linear voltampere characteristics of linear circuits, nonlinear characteristics represent a practical interest too. The above-mentioned
dependence of power via load voltage (1.6) is such nonlinear characteristic. For a
passive two-port network, the dependence of efciency via load voltage is also a
nonlinear function.
Besides linear active two-pole networks, a wide class of active nonlinear
two-poles is known. Solar cells, quasi-resonant voltage converters possess a typical
convex characteristic. Power-source and power-load elements with a concave
characteristic are also known. Further, we will consider features of such nonlinear
characteristics.

1.5.1

Efciency of Two-Ports with Different Losses

Let us consider a two-port in Fig. 1.15. In this case, the power transfer ratio KP or
efciency g is a nonlinear function.
As it is known, the system of equation of this circuit has the view
"

I0
I1

"

Y00  Y10
Y10  Y11

# "

#
V0

;
V1

1:27

18

Fig. 1.15 Two-port with loss


and a variable load
conductivity

V0, I0

Introduction

V1, I1
y10
y0

y1

YL

YIN

where Y parameters
Y00 y10 y0 ;

Y11 y10 y1 ;

Y10 y10 :

The matrix determinant is


DY Y00 Y11  Y10 2 :
The characteristic admittance at the input and output or load is
YIN:C

r
Y00

DY ;
Y11

YL:C

r
Y11

DY :
Y00

1:28

Next, we use the transmission a parameters. Then,


"

V0
I0

"

a11 a12
a21 a22

# "


V1
I1

3
Y11 1
" #
6 Y10 Y10 7 V1
7
6
4 DY Y00 5  I :
1
Y10 Y10
2

1:29

Using the attenuation coefcient c [2], we may rewrite Eq. (1.29) as


2

3

V0
4 I0 5 chc
shc
YIN:C

3
YL:C
V1 p
DY 7
shc 6
7:
6
4
chc
I1 5
p
DY


1:30

In turn, the admittance transformation has the view


YL
thc
YIN
Y
L:C YL
:
YIN:C 1 Y thc
L:C

1:31

1.5 Nonlinear Characteristics

19

We have the relative values YIN =YIN:C ; YL =YL:C . However, expression (1.31) is
not a pure relative because contains the value thc. This value is diverse for
two-ports with different losses.
In turn, the power transfer ratio has the following forms:


8
< KP PP1 ch2 c sh2 c  chc  shc  YYIN YYIN:C
0
IN:C
IN
h

i
: K ch2 c sh2 c chc  shc  YL YL:C 1 :
P
YL:C
YL

1:32

Sequentially, the maximum value KPM corresponds to the admittance matching


KPM ch2 c sh2 c  2 chc  shc chc  shc2 :

1:33

So, the relative values YIN =YIN:C ; YL =YL:C determine the deviation from the
admittance matching by some way. However, expressions (1.32) are not pure relative because they contain the value thc. If we try to use the value KPM , as the scale
value, we cannot obtain the pure relative form of expression (1.32).
Let us use the following voltage transfer ratio:
KG

V1
Y11 V1

;
V1OC Y10 V0

which corresponds to the voltage ratio of the Thvenin equivalent circuit. Using
(1.30), we get
KP KG KG

1  KG
;
A  KG

A ch2 c:

1:34

An example of efciency (1.34) is shown in Fig. 1.16 for the various values
A. The maximum values of KPM correspond to the various values KG .
Fig. 1.16 Efciency of
two-ports for different losses

0.8

KPM
0.6

A=1.05
KP
0.4

KPM

A=1.25
0.2

0.0
0.0

0.2

0.4

0.6

0.8

KG

1.0

20

Introduction

The solid arrows demonstrate the correspondence of the characteristic points. So,
we have the three characteristic regimes. Then, the problem of scales, relative
expressions, and correspondence of running points (dash arrow) arises.

1.5.2

Characteristic Regimes of Solar Cells

Let us consider voltampere and power-volt characteristics of two different solar


~M
cells in Fig. 1.17a. These characteristics have the maximum power points PM ; P
~M [8]. There are also such
with the different currents IM ; ~IM and voltages VM ; V
SC ~SC
~ OC .
characteristic values as SC currents I ; I and OC voltages V OC ; V
The solid arrows show the correspondence of the characteristic points. So, we
have the three characteristic regimes. Similar to Sect. 1.5.1, the problem of scales,
relative expressions, correspondence of running points (dash arrow), and approximation arises [15, 16].

1.5.3

Quasi-resonant Voltage Converter

It is known the load resonant converters, for example, with the zero-current
switching [17]. These converters regulate their output by changing the switching
period TS . Simulation of ORCAD model shows that its load curve is almost

(a)
IL IL
P P

SC

IL

PM

SC

IL

PM

IM

PM

PM
V
0

VM V OC

OC

VL, VL

0
VM V OC

OC

VL,VL

(b)
I
M
I
I
0 V

M
V

P
V

Fig. 1.17 a Voltampere characteristic and power-volt characteristic of solar cells. b Voltampere
characteristic of power-load element

1.5 Nonlinear Characteristics

21

rectangular with maximum load power point PM and similar to Fig. 1.17a. So, we
have the same problems to determine the regime parameters.

1.5.4

Power-Source and Power-Load Elements

The power-source and power-load elements are known from [4, 18]. Their load
characteristics have the typical form as shown in Fig. 1.17b. Analysis of the
power-load element shows the two-valued voltage of a limited capacity voltage
~ But for all that, the voltampere characteristic of this
source; that is, points M; M.
power-load element has one-valued representation.
On the other hand, taking into account the losses of real power-source and
power-load elements, we get the two-valued voltampere characteristic of these
elements. So, we have some characteristic regimes, and the same problems to
determine the regime parameters.

1.6
1.6.1

Regulated Voltage Converters


Voltage Regulator with a Limited Capacity Voltage
Source

In power supply systems with limited capacity voltage sources, the limitation of
load power appears. An independent power supply system with a solar array and
rechargeable battery may be the example of one.
We consider DC transformer with switched tapped secondary windings as a
voltage regulator VR1 connected to a limited capacity voltage source V0 in
Fig. 1.18a. The regulator denes a transformation ratio
n1

V1
:
V

1:35

The regulation characteristic V1 n1 , by an internal resistance Ri , is a nonlinear


and two-valued function shown in Fig. 1.18b. In this case, the upslope direction or
forward branch of its regulation characteristic is used and the down movement of an
operating point on the back branch is restrained.
Let Ri be equal to the load R1 . Then, the expression of V1 n1 has the view
V1
n1
V1
:
V0
1 n1 2

1:36

We get the relative expression which coincides with different circuits. But,
similar to Sect. 1.2.2, the relative regime parameters V 1 ; n1 have the different
quantities. For example, the maximum value V 1M 0:5 corresponds to n1M 1.

22
Fig. 1.18 a Power supply
system with limited capacity
voltage sources. b Example of
its regulation characteristic

Introduction

VR1

(a)

Ri

V1

V0

VR2

n1
V2
R

n2

(b)
0.50

V1
V0
0.25

0.00
0

n1

Therefore, it is necessary to stipulate the parameter to set the regime value in the
form of a number.
Also, there is a question of the next switching step for the secondary winding;
whether the regular or irregular step will be for the load voltage or for transformation ratio relatively to maximum permissible values. It is determined by how to
set the value changes; that is, by increments, ratio and so on. The situation becomes
complicated even more, if such a power supply system contains two loads with
individual voltage regulators shown by dash lines in Fig. 1.18a; the interference of
these loads takes place.

1.6 Regulated Voltage Converters

1.6.2

23

Buck Converter

Let us consider a buck converter in Fig. 1.19.


The expression of the static regulation characteristic for the continuous current
mode of a choke L with a loss resistance R has the known view [5, 10]
VL

V0
V0
D
D;
1 RRL
1 r2

1:37

where D is the relative pulse width and r2 is the relative loss.


Let us express (1.37) in the relative form. There are some variants.
First, it is possible to introduce the value V L VL =V0 . Then
VL

1
1 r2

1:38

D:

This expression is not a pure relative because it contains the value r2 .


Let D be changed, D1 ! D2 , and a subsequent value D2 D1 D21 . Then, the
21
voltage change V L , by (1.38), depends on the parameter r2 .
Let us consider the ratio of the initial and subsequent values
2

VL

1
VL

D2
:
D1

This determination of regime change does not depend on parameter r2 .


Second, we may introduce the value
^L VL D;
V
VLM

1:39

where the maximum load voltage is


VLM

V0
1 r2

L
V0

Fig. 1.19 Buck converter

24

Introduction

Therefore, we obtain the following equalities for regimes of different converters:


D1 D2 ;

^L1 V
^L2 :
V

In turn, the regime changes are expressed by the ratio or increment of values
^L1 ; V
^L2 .
V
If an initial expression of type (1.37) is more complex, it will be also difcult to
constitute relative expression (1.39) directly and reasonably. This case is considered
in the next section.

1.6.3

Boost Converter

Let us consider a boost converter in Fig. 1.20a. Similar to (1.37), we offer an


equation of the known static regulation characteristic
VL
n

;
V0 1 r2 n2

1:40

1
where n 1D
 1 is the inverse relative pause width.

Fig. 1.20 a Boost converter.


b Its regulation characteristics
for different losses

(a)
V0

(b)

2=0.05
VL
V0 2

2=0.1
1

0
1

1.6 Regulated Voltage Converters

25

This expression is analogous to (1.36). The regulation characteristics for various


values r2 are shown in Fig. 1.20b. It is obvious that the maximum values VL
correspond to the various values n. The characteristic value is also n 1.
The solid arrows show the correspondence of the characteristic points and dash
arrow shows running points.
Let us attempt to express (1.40) in the relative form. We may introduce the new
values
^ L r VL ;
V
V0

^n rn:

Then
^L
V

^n
1 ^n2

Although the relative expression turns out, for the characteristic value n 1 (the
transistor is switched off) the new value ^n r depends on the parameter of the
converter.

References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Bessonov L.A.: Teoreticheskie Osnovy Elektrotekhniki. Elektricheskie tsepi, Izd.9. (Basic
electrical engineering theory: Electric circuits, 9th edn.). Vyshaia shkola, Moskva (1996)
3. Bogatyrev, O.M.: General method of solving of problems for linear circuit at changeable
resistances of branches. Elektrichestvo. 9, 6769 (1955)
4. Emadi, A., Khaligh, A., Rivetta, C.H., Williamson, G.A.: Constant power loads and negative
impedance instability in automotive systems: denition, modeling, stability, and control of
power electronic converters and motor drives. IEEE Trans. Veh. Technol. 55(4), 11121125
(2006)
5. Erickson, R.W., Maksimovic, D.: Fundamentals of power electronics. Springer, Berlin (2001)
6. Glisson, T.H.: Introduction to circuit analysis and design. Springer, Berlin (2011)
7. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
8. Ishibashi, K., Kimura, Y., Nivano, M.: An extensively valid and stable method for derivation
of all parameters of a solar cell from a single current-voltage characteristic. J. Appl. Phys. 103
(9), 094507/1094507/6 (2008)
9. IvanovTsyganov, A,I.: Electropreobrazovatelnye ustroistva RES. (Electro-converting
equipments of radio-electronic systems RES). Vyshaia shkola, Moskva (1991)
10. Kazimierczuk, M.K.: Pulsewidth Modulated DCDC Power Converters. Wiley, New York
(2008)
11. Leger, A.S., Nwankpa, C.O.: Analog and hybrid computation approaches for static power
flow. Paper presented at the 40th Annual Hawaii international conference on system sciences
(2007)

26

Introduction

12. Levron, Y., Shmilovitz, D.: On the maximum efciency of systems containing multiple
sources. IEEE Trans. Circuits Syst. I Regul. Pap. 57(8), 22322241 (2010)
13. Liu, X., Wang, P., Loh, P.C.: A hybrid AC/DC microgrid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
14. Nelatury, S.R.: Didactic uses of maximum power transfer theorem and guided problem
solving. Int. J. Electr. Eng. Educ. 51(3), 244260 (2014)
15. Penin, A., Sidorenko, A.: A convenient model for I-V characteristic of a solar cell generator as
an active twopole with selflimitation of current. World Acad. Sci. Eng. Technol. 3(4),
905-909 (2009). http://www.waset.org/publications/9926. Accessed 30Nov 2014
16. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010). http://sfm.asm.md/
moldphys/2010/vol9/n2/index.html. Accessed 30 Nov 2014
17. Penin, A.: A quasi-resonance voltage converter with improved parameters. Elektrichestvo. 2,
5864 (2009)
18. Singer, S., Erickson, R.W.: Powersource element and its properties. IEEE Proc. Circ. Devices
Syst. 141(3), 220226 (1994)
19. Venikov, V.A.: Theory of similarity and simulation: with applications to problems in electrical
power engineering. Macdonald, London (1969)

Part I

Electrical Circuits with one Load.


Projective Coordinates of a Straight
Line Point

Chapter 2

Operating Regimes of an Active Two-Pole.


Display of Projective Geometry

2.1

VoltAmpere Characteristics of an Active Two-Pole.


Afne and Projective Transformations of Regime
Parameters

2.1.1

Afne Transformations

It is known that any linear circuit (an active two-pole A) relative to load terminals is
replaced by a voltage source V0 in series with an internal resistance Ri [1, 6]. Let us
consider two cases of the load.
Case 1
A variable voltage source VL is the load of an active two-pole A in Fig. 2.1. The
voltage VL (independent quantity) and load current IL are the parameters of operating or running regime.
At change of the load voltage from the short circuit SC VL 0 to open circuit
OC VLOC V0 , a load straight line is given by (1.1)
IL

V0 VL
VL

ILSC  ;
Ri Ri
Ri

2:1

where ILSC is the SC current. This straight line is shown in Fig. 2.2.
The characteristic regime parameters ILSC ; V0 determine the different scales for
the axes of coordinates.
We want to represent a running regime parameter of the load by a certain
quantity, which would have an identical value for various actual regime parameters
such as voltage and current. To do this, we use a geometrical interpretation of
change or kinematics of a circuit regime.
As the load characteristic is dened by linear expression (2.1), a similar
expression in geometry denes an afne transformation, conformity, or mapping of
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_2

29

2 Operating Regimes of an Active Two-Pole

30
Fig. 2.1 Active two-pole
with a load voltage source

Fig. 2.2 Load straight line of


an active two-pole

the voltage axis to current axis VL ! IL [7]. The values ILSC ; Ri are parameters of
this afne transformation. The mechanism of this mapping is shown by parallel
lines with arrows for three voltage values or points VL1 ; VL2 ; VL3 .
An afne transformation is characterized by an invariant of three points
uVL1 ; VL2 ; VL3 uIL1 ; IL2 ; IL3 :
The invariant represents a certain expression, which only contains the chosen
values of voltage or corresponding values of current; the parameters of the afne
transformation do not enter in this expression.
To obtain the invariant expression, it is necessary to exclude the two parameters
from Eq. (2.1) by means of three equations. For arbitrary three voltage values
VL1 ; VL2 ; VL3 we have the system of equations

IL1 ILSC 

VL1
;
Ri

IL2 ILSC 

VL2
;
Ri

IL3 ILSC 

VL3
:
Ri

2:2

Using the rst and second equations, we exclude ILSC


IL1  IL2

VL2 VL1
 :
Ri
Ri

2:3

2.1 VoltAmpere Characteristics of an Active Two-Pole

31

Similarly, we have
IL1  IL3

VL3 VL1
 :
Ri
Ri

2:4

IL2  IL3

VL3 VL2
 :
Ri
Ri

2:5

Let us exclude the parameter Ri . To do this, we use expressions (2.3) and (2.5).
Then, the required invariant or afne ratio is obtained as follows:
IL1  IL2 VL2  VL1

:
IL2  IL3 VL3  VL2

2:6

If we use expressions (2.3) and (2.4), we get the other invariant


IL1  IL3 VL3  VL1

:
IL1  IL2 VL2  VL1

2:7

Application of the concrete invariant is determined by a physical sense of a


parameter of running regime, as it will be shown later.
Let us consider expression (2.6). We accept values IL1 ; VL1 corresponding to the
running regime. In turn, values IL2 ; VL2 correspond to the open-circuit regime,
IL2 0; VL2 V0 ; values IL3 ; VL3 correspond to the short circuit regime,
IL3 ILSC ; VL3 0. The pairs of the respective points IL2 ; VL2 and IL3 ; VL3 are the base
or extreme points; the points IL1 ; VL1 are the dividing points.
Then, expression (2.6) takes the form
IL1  0
V0  VL1

;
0  V0
0  ILSC
IL1  0
V0  VL1
:

V0  0
ILSC  0

2:8

The afne ratio n1 can be represented by the formal view


n1 IL1 0 ILSC VL1 V0 0;
where
IL1 0 ILSC

0  IL1
IL1

;
0  ILSC ILSC

2:9

2 Operating Regimes of an Active Two-Pole

32

Fig. 2.3 Afne transformation VL ! IL

VL1 V0 0

V0  VL1
:
V0  0

2:10

Further, we consider the sense of the value n1 . We obtain the quantity that has an
identical value for current and voltage. For the current, this value n1 is simply the
normalized value (at the expense of a particular choice of the base points).
Therefore, expression (2.8) coincides with Eq. (2.1) for the normalized values
IL1
V1
1 L:
SC
V0
IL

2:11

The more convenient representation of afne transformation (2.1) is given in


Fig. 2.3 for the actual values of the current and voltage. There is a projection center
S and straight lines VL ; IL are parallel. The afne conformity is also given by two
pairs of respective base points.
As it was already shown, as the respective base points, it is convenient to use the
points of characteristic regimes, which can be easily determined at a qualitative
level; that is, the short circuit and open-circuit regimes. The sense of this afne
transformation is visible in Fig. 2.3. The projection center S has a nal coordinate
because of different scales or base points of the current and voltage lines.
If the projection center is S ! 1, the projection is carried out by parallel lines.
This mapping corresponds to the Euclidean transformation; that is, the parallel
translation of a segment in Fig. 2.4. Such a case conforms to Eq. (2.11).
Now, we consider changes of regime; that is, how to set this change in the
invariant form too. Let an initial regime be given by values IL1 ; VL1 and subsequent
regime be set by IL2 ; VL2 . Let us express the subsequent value of current through the
initial value. Using Eq. (2.3), we get
IL2 IL1

VL1 VL2

IL1 IL21 :
Ri
Ri

2:12

The obtained transformation with a parameter IL21 translates the initial regime IL1
point into the subsequent regime IL2 point; that is, IL1 ! IL2 shown in Fig. 2.5.

2.1 VoltAmpere Characteristics of an Active Two-Pole

33

Fig. 2.4 Euclidean transformation VL ! IL

Fig. 2.5 Translation of point IL1 ! IL2 ! IL3 along a line

Fig. 2.6 Movement of a segment I21

Further, transformation (2.12) with a parameter IL32 translates the point IL2 ! IL3 ;
that is,
IL3 IL2 IL32 :
For example, let IL32 be equal IL21 , as it is shown in Fig. 2.5. It now follows that
IL3 IL1 IL21 IL32 IL1 IL31 :
Therefore, we have the resultant transformation IL1 ! IL3 with a parameter IL31 .
Thus, a set of transformations (2.12) is a group.
Let an initial regime be given by two values I11 ; I21 . These values correspond to a
segment I21 in Fig. 2.6.

2 Operating Regimes of an Active Two-Pole

34

Fig. 2.7 Signal V21 and bias VL voltage is the load of an active two-pole

For example, this segment or increment of current I21 corresponds to a signal


voltage V21 in series with a variable bias voltage VL in Fig. 2.7.
If we apply transformation (2.12) to the initial points I11 ; I21 , we obtain the
subsequent points
I12 I11 I 21 ;

I22 I21 I 21 :

2:13

Obviously, translation (2.12) or (2.13) is characterized by the invariant of two


points uI11 ; I21 I21  I11 , because
I22  I12 I21  I11 :

2:14

Thus, this invariant is the Euclidean distance between two points. Therefore,
geometry of this transformation group is geometry of a Euclidean straight line.
Also, we can consider the dual variant of segment and its movement in Fig. 2.8.
Let an initial regime be given by two values I11 ; I12 or segment I12 I11 . We use the
translation with a parameter I21 . Then, the subsequent points
I21 I11 I21 ;

Fig. 2.8 Movement of segment I12 I11 ! I22 I21

I22 I21 I21 :

2.1 VoltAmpere Characteristics of an Active Two-Pole

35

Obviously, there is an invariant of two points


I12  I11 I22  I21 :
Using (2.11), we get the similar invariant for the normalized voltage and current
values
IL2
IL1
VL1 VL2



n21 :
ILSC ILSC V0 V0

2:15

This regime change corresponds to a relative change in percent. Note that the
invariants of transformation (2.12) coincide with the known principle of superposition for linear circuits.
Remark Let us introduce a regime change by ratio of currents. Using expression
(2.11), we get
IL2
IL1
IL2 V0  VL2


:
ILSC ILSC IL1 V0  VL1
We can note that the ratio of currents does not contain a transformation
parameter. But, this ratio of voltages contains the transformation parameter V0 .
Hence, the condition of invariant is not executed.
Case 2
A variable conductivity YL is a load of the simplest circuit in Fig. 2.9; an internal
resistance of this active two-pole Ri 0.The equation of this circuit
IL V0 YL :

2:16

The conductivity YL (independent quantity) and load current are the parameters
of a running regime.

Fig. 2.9 Active two-pole with Ri 0

2 Operating Regimes of an Active Two-Pole

36

Fig. 2.10 Transformation of a point IL1 ! IL2 ! IL3

For this circuit, the regime has only an absolute value; it is impossible to state a
relative expression in view of the absence of a scale.
Now, we consider changes of regime. The initial conductivity value equals YL1
and subsequent one equals YL2 . Then, we have the two values of currents
IL1 V0 YL1 ;

IL2 V0 YL2 :

2:17

Using the ratio of these equations, we get


IL2 YL2

m21 :
IL1 YL1

2:18

Then, the subsequent current is obtained as


IL2 m21 IL1 :

2:19

This transformation with a parameter m21 translates a point of an initial regime IL1
to a point of a subsequent regime IL2 ; that is, IL1 ! IL2 . The transformation is a group
transformation because
IL3 m32 IL2 m32 m21 IL1 m31 IL1 :
This group is the dilation group. For example, let m32 be equal m21 , as shown in
Fig. 2.10; a usual distance between currents is increased.
Let an initial regime be given by two values I11 ; I21 . These values correspond to
segment I1 I2 in Fig. 2.11. If we apply transformation (2.19) to the initial points
I11 ; I21 or to the segment, we obtain the subsequent points

Fig. 2.11 Movement of segment I1 I2

2.1 VoltAmpere Characteristics of an Active Two-Pole

I12 m21 I11 ;

37

I22 m21 I21 :

Obviously, transformation (2.19) has the invariant of two points


uI22 ; I12

I22 m21 I21 I21

m21 uI21 ; I11 :


I12 m21 I11 I11

2:20

This invariant m21 is the invariable length of segment I1 I2 . Therefore, we


obtain the segment end
I21 m21 I11 :
These two examined cases correspond to the known similarity property.
And so, the consideration of regime changes as geometrical transformations
gives a methodical foundation for the valid introduction of regime changes for more
complex cases.

2.1.2

Projective Transformations

Let us consider a circuit with a variable load resistance in Fig. 2.12.


The load straight line is also given by expression (2.1)
IL

V0 VL
VL

ILSC  :
Ri Ri
Ri

This straight line is shown in Fig. 2.13.

Fig. 2.12 Circuit with a load


resistance

2:21

38

2 Operating Regimes of an Active Two-Pole

Fig. 2.13 Load straight line


of a circuit with a load
resistance

The bunch of straight lines with a parameter RL corresponds to this straight line.
The bunch center is the point 0. The equation of this bunch is given as
IL

1
VL :
RL

2:22

Further, it is possible to calibrate the load straight line in the load resistance
values [10, 13, 15]. The internal area RL [ 0 at the load change corresponds to
regime of energy consumption by the load. If to continue the calibration for the
negative values RL \0, the regime passes into the external area, which physically
means return of energy to the voltage source V0 .
Therefore, at the innitely remote point RL Ri , the calibrations of the load
straight line will coincide for the areas VL [ V0 ; VL \0. So, this straight line is
closed; that is typical property of projective geometry.
The load resistance value determines the nonhomogeneous coordinate for a point
of load straight line. In turn, the two values VL ; IL are the homogeneous coordinates
because RL qVL  qIL , where q is any nonzero real number. Homogeneous
coordinates have nite values. Further, the equation VL RL has the characteristic
fractionally linear view
VL V0

RL
:
Ri RL

2:23

Of course, similar expressions take place for currents and resistances for different
branches of active circuits and are used for measurement of circuit parameters [2].
That gives the solid grounds for considering the map RL ! VL as a projective
transformation of projective geometry [4, 5, 7]. In general, a projective transformation of points of one straight line into points of another line is set by a center S of
projection or three pairs of respective points in Fig. 2.14. Therefore, the innitely
remote point RL 1 corresponds to the nite point VL V0 . Also, fractionally
linear expressions of type (2.23) as projective transformations are used for

2.1 VoltAmpere Characteristics of an Active Two-Pole

39

Fig. 2.14 Projective


transformation of points
RL ! VL

V0

V0
2

VL

VL

RL

RL

Ri

mL

VL

21

mL

mL

RL

mL

measuring instruments [8, 9]. In addition, a class of innite ideal points of projective geometry is introduced to represent innite resistances [3].
As the pairs of respective points, it is convenient to use the points of the
characteristic regimes, which can be easily determined at a qualitative level; that is,
the short circuit, open circuit, and maximum load power. In turn, the point of a
running regime is the fourth point.
We want to represent the running regime of load in the relative form regarding to
these three characteristic or base points. To do this, we may use a cross-ratio of four
points.
The projective transformations preserve a cross-ratio of four points. For values
R1L ; VL1 ; IL1 of initial or running regime, the cross-ratio mL has the view
m1L 0 R1L Ri 1
m1L 0 VL1

m1L

ILSC

IL1

R1L  0 Ri  0
R1L

;
:
R1L  1 Ri  1 Ri

V0
0
V0
V1  0
VL1
V0 1L
 V02

;
2
VL  V0 2  V0 V0  VL1

ILSC
I 1  I SC
0 L 1 L 
2
IL  0

ILSC
SC
2  IL
SC
IL
2 0

ILSC  IL1
:
IL1

2:24
2:25

2:26

We note that expression (2.26) corresponds to (2.7).


The cross-ratio in geometry underlies the denition of the distance between
points R1L ; RL Ri concerning the extreme or base values 0; 1. The point Ri is a
scale or a unit point. Similarly, we have the base points 0; V0 , and a unit point V0 =2
of voltage. Thus, a projective coordinate of running regime point is set by the value
mL , which is dened in an identical (invariant) manner through various regime
parameters as RL ; VL ; IL .
In turn, a regime change R1L ! R2L (respectively VL1 ! VL2 , IL1 ! IL2 ) can be
expressed similarly
2 1
m21
L 0 RL RL 1

R2L
;
R1L

2:27

2 Operating Regimes of an Active Two-Pole

40

2
1
m21
L 0 VL VL V0

VL2
VL1

;
V0  VL2 V0  VL1

2:28

SC 2 1
m21
L IL IL IL 0

ILSC  IL2 ILSC  IL1



:
IL2
IL1

2:29

The cross-ratio has the next quality


1 2
m12
L 0 RL RL 1

R1L
1
21 :
2
RL mL

A group property of the cross-ratio realizes


1
m2L m21
L mL :

2:30

For the next regime change R2L ! R3L , the group property is given by
2
32 21 1
31 1
m3L m32
L mL mL mL mL mL mL :

Let us express the subsequent voltage value VL2 by the initial value VL1 and
regime change value m21
L . Using (2.28), we get the recalculation formula
V1

21
L
VL2
V0 mL
V1
:
V0
L
m21  1 1
V0

2:31

This formula allows nding a subsequent voltage value by an initial voltage and
transformation parameter m21
L . Also, this expression can be obtained from (2.23).
For an initial R1L and subsequent R2L value, we get the following system of the
equations:
8
1
< V 1 V0 RL 1
L
Ri R

: V 2 V0 RL 2
L
Ri R
2

Excepting Ri , we get expression (2.31).


In general, transformation (2.31) translates an initial point VL1 to subsequent
point VL2 . Therefore, we can set the identical regime changes for the different initial
regimes shown in Fig. 2.15 for the straight line VL as a closed projective straight
line.
The identical transformation parameter m21
L forms a segment of invariable
length (in sense of projective geometry) and we observe the movement of this
segment. Here, the change of the Euclidean (usual) length is visible. For the base
points, the Euclidean length is decreasing to zero.

2.1 VoltAmpere Characteristics of an Active Two-Pole

41

Fig. 2.15 Identical regime changes for the different initial regimes

In the theory of the projective transformations, the xed points play an important
role. For their nding, Eq. (2.31) is solved as VL1 VL2 . It turns out the two real
roots, VL 0; VL V0 , which dene the hyperbolic transformation of hyperbolic
(Lobachevski) geometry. Physically, the xed point means such a regime when a
variable VL does not depend on the initial or subsequent value RL . It is evident for
SC; OC regimes.
If the roots of equation coincide, the xed point denes the parabolic transformation and, respectively, parabolic (Euclidean) geometry. If the roots are imaginary, geometry is elliptic (Riemannian).
In geometry, it is established that these three kinds of transformations (projective, afne, and Euclidean) exhaust possible variants of group transformations,
which underlie the denition of the metrics of a straight line. Thus, the geometrical
approach allows validating regimes determination, and both denitions of a regime
and its change are coordinated by structure of expressions and ensure the performance of group properties.
Reasoning from such a geometrical interpretation, it is possible to give the
following denition [12, 13]:
a circuit regime is a coordinate of point on load straight lines and axes of
coordinates;
a regime change is a movement of point on all the straight lines, which denes a
segment of corresponding length.
In this connection, it is possible to accept the following requirement (likewise to
metric space axioms):
independence or invariance of regimes and their changes from variables (regime
parameters) as type R; V; I;
the additive postulate of regimes changes;
assignment of equal regime change for various initial regimes.

2 Operating Regimes of an Active Two-Pole

42

2.2

VoltAmpere Characteristics of an Active Two-Pole


with a Variable Element

2.2.1

Thvenin Equivalent Circuit with the Variable


Internal Resistance

Let us consider the Thvenin equivalent circuit with the variable internal resistance
Ri in Fig. 2.16.
In this case, a bunch of load straight lines with a parameter Ri is obtained at a
center G in Fig. 2.17. The unied equation of this bunch is given by
IL

1
VL V0 :
Ri

2:32

The coordinate of the center G corresponding to V0 does not depend on Ri .


Physically, it means that the current across this element is equal to zero. The
element Ri can accept the two base or characteristic values, as 0; 1. The third
characteristic value is not present for Ri .
Also, the bunch of straight lines with a parameter RL corresponds to these load
straight lines. The point 0 is the bunch center. Therefore, RL can accept the two base
or characteristic values, as 0; 1. Hereafter, we say about the two bunches with the
parameter Ri and RL correspondingly.
Let relative regimes be considered for this circuit.
Case 1
Let the internal resistance Ri be equal to R1i and the load resistance varies from R1L to
R2L . In this case, a point of initial regime is C1 ! C2 . If Ri is equal to R2i , a point of
initial regime B1 ! B2 . Therefore, the regime change, which is determined by the
load change (an own change), is expressed similar to (2.27)

Fig. 2.16 Thvenin


equivalent with the variable
internal resistance

2.2 VoltAmpere Characteristics of an Active Two-Pole with a Variable Element

43

Fig. 2.17 Bunch of load


straight lines for a voltage
source

m21
L CI C2 C1 G BI B2 B1 G
0 R2L R1L 1

R2L
:
R1L

2:33

This determination of a regime change does not depend on Ri . Therefore, we


must use only the load voltage for this calculation, as (2.28)
C2
C1
m21
L 0 VL VL V0

VLC2  0
VLC1  0
 C1
:
C2
VL  V0 VL  V0

2:34

In this case, the base points CI ; BI correspond to the common value VL 0.


Case 2
Similarly, the regime change, which is determined by Ri change (a mutual change),
is given by
m21
i 0 B2 C2 A2 0 B1 C1 A1
1 R2i R1i 0

R1i
:
R2i

2:35

This determination of a regime change does not depend on RL . Therefore, we


must use only load voltage for calculation, as (2.34)
B2
C2
m21
i 0 VL VL V0

VLB2  0
VLC2  0
 C2
:
B2
VL  V0 VL  V0

2:36

In expressions (2.34) and (2.36) the identical base points, which are the centers
of the two bunches of straight lines RL ,Ri , are used.

2 Operating Regimes of an Active Two-Pole

44

Case 3
If the regime is changed as point C1 ! C2 ! B2 , it is possible to speak about the
general or compound change
21
m21 m21
i mL

R1i R2L
:
R2i R1L

2:37

Then

21

VLB2  0
VLC2  0

VLB2  V0 VLC2  V0

  C2

VL  0
VLC1  0


:
VLC2  V0 VLC1  V0

Finally, we obtain
m21

VLB2  0
VLC1  0

0 VLB2 VLC1 V0 :
VLB2  V0 VLC1  V0

2:38

Analogous to (2.31), we get the recalculation formula


V C1

21
L
VLB2
V0 m
V C1
:
V0
L
m21  1 1

2:39

V0

The compound regime and voltage changes, as point moving, are shown in
Fig. 2.18.
We may note that values RL , Ri have not any scales.

2.2.2

Norton Equivalent Circuit with the Variable Internal


Conductivity

Let us consider the Norton equivalent circuit with the variable internal resistance or
conductance Yi and load conductance YL in Fig. 2.19.

Fig. 2.18 Compound voltage changes

2.2 VoltAmpere Characteristics of an Active Two-Pole with a Variable Element

45

Fig. 2.19 Norton equivalent with the variable internal conductance


Fig. 2.20 Bunch of load
straight lines for a current
source

In this case, a bunch of load straight lines Yi is obtained at the center G in


Fig. 2.20.
The unied equation of this bunch is given as
IL  I0 Yi VL :

2:40

The coordinate of the center G corresponding to I0 ILSC does not depend on the
value Yi . Physically, it means that the current across this element is equal to zero.
The element Yi can accept the two base or characteristic values, as 0; 1. The third
characteristic value is not present for Yi .
Let the relative regimes be considered for this case.
Case 1
Let the internal conductance Yi be equal to Yi1 and the load conductance varies from
YL1 to YL2 . In this case, a point of initial regime C1 ! C2 . If Yi is equal to Yi2 , a point
of initial regime B1 ! B2 . Therefore, the regime change, which is determined by
the load change (an own change), is expressed similarly to (2.33)
m21
L CV C2 C1 G BV B2 B1 G
0 YL2 YL1 1

YL2
:
YL1

2:41

2 Operating Regimes of an Active Two-Pole

46

This determination of a regime change does not depend on Yi . Therefore, we


must use only the load current for calculation; that is,
C2 C1
m21
L 0 IL IL I0

ILC2  0
ILC1  0

:
ILC2  V0 ILC1  I0

2:42

In this case, the base points CV ; BV correspond to the common value IL 0


Case 2
Similarly, the regime change, which is determined by Yi change (a mutual change),
is given as
m21
i 0 B2 C2 F2 0 B1 C1 F1
1 Yi2 Yi1 0

Yi1
:
Yi2

2:43

This determination of a regime change does not depend on YL . Therefore, we


must use only the load current for calculation
B2 C2
m21
i 0 IL IL I0

ILB2  0 ILC2  0

:
ILB2  I0 ILC2  I0

2:44

Case 3
If regime is changed as C1 ! C2 ! B2 , the compound change is
21
m21 m21
i mL

Yi1 YL2
:
Yi2 YL1

2:45

Then
m21

ILB2  0 ILC1  0

0 ILB2 ILC1 I0 :
ILB2  I0 ILC1  I0

2:46

The recalculation formula


I C1

21
L
ILB2
I0 m
I C1
:
I0
L
m21  1 1
I0

The compound regime and current changes are shown in Fig. 2.21.

2:47

2.2 VoltAmpere Characteristics of an Active Two-Pole with a Variable Element

47

Fig. 2.21 Compound current changes

We may note that values YL , Yi have not any scales also.


The above-mentioned arguments make it possible to confront regimes of compared circuits and give the basis for analysis of the general case of circuit.

2.3

Regime Symmetry for a Load-Power

In the above Sect. 2.1, the examined regime parameters of the type V; I; R are
expressed among themselves by linear and fractionally linear expressions (2.1),
(2.23) and have projective properties. In turn, such an important energy characteristic, as a load-power, represents quadratic expression (1.6) and determines a
parabola in Fig. 1.6. This quadratic curve has similar projective properties that
permit to compare the regime of different circuits, to determine the deviation from
the power matching [11, 12, 14]. Let us consider these properties in detail.
To do this, we use the circuit in Fig. 2.12 and rewrite Eq. (2.21) of load straight
line in the following relative form
IL
VL
I 1
1  KV ;
V0
ILSC

2:48

where KV is the voltage transfer ratio. Also, we rewrite Eq. (1.6) of load power in
the similar relative form
P

PL
KV  KV 2 ;
PSC
0

2:49

where PSC
0 is the maximum power of the voltage source V0 for SC regime.
Therefore, the task of equal regimes does not cause a problem; that is simply
corresponding equality of values KV ; I; P. But a deeper analysis will be useful,
which allows generalizing the justication of the equality of regimes and will be
used for considering a more complex case, the efciency of two-ports.

2 Operating Regimes of an Active Two-Pole

48

2.3.1

Symmetry of Consumption and Return of Power

Let us consider load straight line (2.48) in Fig. 2.22.


In the rst quadrant, a positive load consumes energy; there is the maximum
load power point PLM for RL Ri . At SC and OC regime points, RL 0; 1, load
power (2.49) is equal to zero, as shown in Fig. 2.23.
We remind if a negative load returns energy into the voltage source V0 , then the
load voltage VL \0 and VL [ V0 . In this case, the load resistance RL \0; the load
power increases. In the nal analysis, the load power P
L M 1 for the resistance
RL Ri . Therefore, the load straight line is closed and the parabola is a closed
oval curve too, which concerns the innitely remote straight line TP
LM at the point
P
L M.
Let us use expression (2.24) of the cross-ratio
m1L 0 R1L Ri 1

R1L
:
Ri

2:50

For the point RL Ri , cross-ratio (2.50)


mL Ri

Ri
1:
Ri

This special case of the fourth point Ri determines the property of the harmonic
conjugacy of four points, which determines the symmetry of points Ri , Ri relatively to base points 0, .

Fig. 2.22 Conformity of


points for positive and
negative load

2.3 Regime Symmetry for a Load Power


Fig. 2.23 a Parabola of
power for the Cartesian
coordinates, b this parabola
into the projective coordinates
is as a closed curve

49

(a)

(b)

For an arbitrary running load RL , we obtain, at once, the corresponding conjugate load resistance equals RL because cross-ratio (2.50)
0  RL RL 1 1:

2:51

So, we have the symmetry of points RL , RL relatively to the base points 0,
too. The points Ri ; Ri ; R1L ; R1L and so on pass into each other, as it is shown by
arrows. Physically, this symmetry corresponds to mapping of the region of power
consumption by a load on the region of return. Then, the mapping of points of
parabola of the region P [ 0 onto the region P\0 is realized from a point F. In this
case, the points 0, 1 of the axis KV are xed. Therefore, we obtain the following
condition:
0 KV RL KV RL 1

KV RL  0 KV RL  0

1:
KV RL  1 KV RL  1

2:52

2 Operating Regimes of an Active Two-Pole

50

From here, we get


KV RL  KV RL
0:5:
KV RL KV RL

2:53

The point F is formed due to the intersection of the tangential FX; FY at the xed
points. This point F is called a pole, and a straight line, passing through the xed
points 0, 1 is a polar 0T. The indicated symmetry is obtained relatively to the polar.
We may pass to the projective system of coordinates YFX. In this coordinate
system, a polar is considered as the innitely remote straight line. Therefore, our
initial parabola will be already a hyperbola, and the coordinate axes FX, FY are
asymptotes in Fig. 2.24. The point P
L M has a nite value.
In this case, a point on hyperbola is assigned as the rotation of radius-vector RF F
from the initial position at the point PLM .
The non-Euclidean distance R1 P
M is determined by a hyperbolic arc length of a
hyperbola; it will be later on shown in Sect. 4.4.

2.3.2

Symmetry Relatively to the Maximum Power Point

Also, the symmetry of points R1L ; R2L , as the points of equal power, is manifested by
arrows relatively to the point Ri or relatively to the straight line PLM P
L M in
Figs. 2.25 and 2.26. The points Ri ; Ri are the xed points. This symmetry
corresponds to points KV1 , KV2 also.
Using (2.49), we get the following condition:
KV1 KV2 1:

2:54

Also, the mapping of points R1L ! R2L , relatively to the straight P


M ; PM of the
parabola, leads to an additional system of pole and polar; the point T is a pole, and

the straight line P


M PM is a polar. Similarly, the regime or the point of the parabola,

Fig. 2.24 Rotation of a


radius-vector RF F

2.3 Regime Symmetry for a Load Power

51

Fig. 2.25 Mapping of points


relatively to the maximum
load power point

Fig. 2.26 Mapping of a


parabola relatively to the
maximum load power line.
a Cartesian coordinates,
b projective coordinates

(a)

(b)

2 Operating Regimes of an Active Two-Pole

52

is assigned as the rotation of radius-vector RT T from the initial position at the point
0 or point 1. In the Cartesian coordinate system, this radius-vector RT T is the
parallel line to the axis KV .
Similar to (2.50), we introduce the other value of running regime
m1L Ri R1L 0  Ri

R1L  Ri 0  Ri R1L Ri

1
:
R1L Ri 0 Ri
RL Ri

2:55

The point RL 0 is a unit point.


Similar to (2.51), we obtain the harmonic conjugate point using the condition
Ri R1L R2L  Ri

R1L  Ri R2L  Ri

1:
R1L Ri R2L Ri

From that, we get


R1L R2L Ri 2 :

2.3.3

2:56

Two Systems of Characteristic Points

Thus, we have obtained a kinematics diagram of the regime deviations relatively


to the selected base points and initial point. Also, we get two conjugate systems of
pole and polar. For example, there are four harmonic conjugate points 0; Q; 1; T
onto the polar 0T of the pole F. Reciprocally, there are four harmonic conjugate


points P
M ; Q; PM ; F onto the polar PM PM of the pole T. Let us consider this
harmonic conjugacy in detail.
For the pole T, the following correspondences take place. We believe the harmonic conjugate points 0; 1 relatively to the base points Q; T of the polar 0T. Then,
these points correspond to four points 0; 1; KV Q 0:5; KV T 1 of the axis
KV . The mutual mapping of the points 0; 1 relatively to xed points KV T; KV Q
is shown by arrows in Fig. 2.27.

Fig. 2.27 Mutual mapping of points relatively to the xed points onto the straight line TR1

2.3 Regime Symmetry for a Load Power

53

In turn, there are such harmonic conjugate points R1L ; R2L relatively to the base
points A; T of running straight line TR1 . The harmonic conjugate points
KV1 ; KV2 ; 0:5; 1 of the axis KV correspond to these points of the line TR1 .
The mutual mapping of points KV1 ; KV2 relatively to the xed or base points
1; 0:5 is shown by arrows in Fig. 2.27 too. Therefore, we can constitute the
cross-ratio
0:5 KV1 KV2 1

KV1  0:5 KV2  0:5 KV1  0:5




1:
KV1  1 KV2  1 KV2  0:5

From this, we get expression (2.54) that conrms the accepted geometrical model.
Similarly, for the pole F, the following correspondences take place. We believe

the harmonic conjugate points PLM


; P
LM relatively to the base points F; Q of the

polar PLM PLM . In turn, there are such harmonic conjugate points R1L ; R1L relatively
to the base points F, C of running straight line FR1 . The harmonic conjugate points
KV C; KV R1 ; 0:5; KV R1 of the axis KV correspond to these points of the line
FR1 .
The mutual mapping of points KV R1L ; KV R1L relatively to the points
KV C; 0:5 is shown by arrows in Fig. 2.28.
Now, it is necessary to nd the common xed or base points for these two
systems of harmonic conjugate points.
We note at once that points KV R1L ; KV R1L are the harmonic conjugate points
relatively to the points 0, 1 in accordance with (2.52). Obviously, the points T, Q
are the harmonic conjugate points relatively to the points 0, 1 too. These correspondences are shown by dash arrows in Fig. 2.29. Therefore, we may choose the
points 0, 1 as the base points.
Thus, the suggested geometrical approach allows to reunite all the points of
characteristic regimes into one system and to prove the choice of base points and a
unit point. Therefore, it is possible to express any running regime point by a
cross-ratio of type (2.50). All the values of cross-ratios for the characteristic and
running points are shown on the axis m in Fig. 2.29.

Fig. 2.28 Mutual mapping of points of the straight line FR1

54

2 Operating Regimes of an Active Two-Pole

Fig. 2.29 Mutual mapping of


points relative to the xed
points and correspondence of
the different values

References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGraw-Hill,
New York (2009)
2. Bhattacharyya, S.P., Keel, L.H., Mohsenizadeh, D.N.: Linear Systems: A Measurement Based
Approach. Springer, India (2014)
3. Bryant, R.E., Tygar, J.D., Huang, L.P.: Geometric characterization of series-parallel variable
resistor networks. Circ. Syst. I Fundam. Theory Appl. IEEE Trans. 41(11), 686698 (1994)
4. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw-Hill,
New York (1967)
5. Glagolev, N.A.: Proektivnaia geometria (Projective Geometry). Nauka, Moskva (1963)
6. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
7. Kagan, V.F.: Osnovania geometrii, Chasti II (Geometry Basics. Part II). Gostekhizdat, Moskva
(1956)
8. Mazin, V.D.: Method for raising the precision of measuring instruments and transducers.
Meas. Tech. 23(6), 479480 (1980)
9. Mazin, V.D.: Error of measurement in the compound-ratio method. Meas. Tech. 26(8), 628
629 (1983)
10. Penin, A.: Projectively-afne properties of resistive two-ports with a variable load.
Tehnicheskaia elektrodinamika 2, 3842 (1991)
11. Penin, A.: Denition of deviation from the matching regime for two-port circuit.
Electrichestvo 4, 3240 (1994)
12. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
13. Penin, A.: Determination of regimes of the equivalent generator based on projective geometry:
the generalized equivalent generator. World Acad. Sci. Eng. Technol. 2(10), 703711 (2008).
http://www.waset.org/publications/10252. Accessed 30 Nov 2014
14. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013). https://
sites.google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 November 2014
15. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010). http://sfm.asm.md/
moldphys/2010/vol9/n2/index.html. Accessed 30 November 2014

Chapter 3

Generalized Equivalent Circuit


of an Active Two-Pole with a Variable
Element

3.1

Introduction

To simplify the calculation of circuits with variable parameters of elements,


Thvenin/Helmholtz and Norton/Mayer theorems are used [1, 68]. In practice, it
can be DC power supply systems with variable loads.
According to these theorems, the xed part of a circuit, concerning terminals of
the specied load, is replaced by an equivalent circuit or equivalent generator. The
open circuit voltage (or short current) and internal resistance are the parameters of
this equivalent generator. These parameters can be used as scales for the normalized
values of load parameters or load regimes. Such a denition of the relative regimes
allows comparing or setting the regimes of different systems.
Considering importance of ideas of the equivalent generator, the attention is
given to the respective theorems in education [2, 14]. Also, these theorems attract
the attention of researchers [35].
However, this known equivalent generator does not completely disclose the
property of a circuit, for example, power supply systems with a base (priority) load
and a variable auxiliary (buffer) load or voltage regulator. In this case, the change of
the auxiliary load leads to change of the open circuit voltage and short circuit
current, as the parameters of the equivalent generator. Thus, the problem of the
calculation of this circuit and nding of the equivalent generator parameters again
arises.
In this chapter, the generalized equivalent generator, which develops Thvenin/
Helmholtz and Norton/Mayer theorems, is proposed. It appears that the load
straight line at various values of anyone changeable resistance passes into a bunch
of these lines. Since the bunch center coordinates do not depend on this changeable
element, then these coordinates can be accepted as the parameters of the generalized
equivalent generator [911]. Also, the approach based on projective geometry of
Chap. 2 for interpretation of changes (kinematics) of regimes is developed [12].
That allows revealing the invariant properties of a circuit; that is, such expressions,
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_3

55

3 Generalized Equivalent Circuit of an Active Two-Pole

56

which turn out identical to the load and element changes. Such invariant expressions permit to obtain convenient formulas of recalculation of the load current.

3.2
3.2.1

Circuit with a Series Variable Resistance


Disadvantage of the Known Equivalent Circuit

Let us consider an active two-pole circuit with a variable series resistance r0N and
base load resistance RL in Fig. 3.1. This circuit has a practical importance for a
voltage regulation.
At change of the load resistance from the short circuit SC to open circuit OC for
the specied series resistance r0N, a load straight line is given by
IL

VLOC VL
VL

ILSC  ;
Ri
Ri
Ri

3:1

where ILSC is the SC current; the OC voltage VLOC and the internal resistance Ri are
the parameters of the Thvenin equivalent circuit in Fig. 3.2.
For our two-pole circuit, we have
VLOC V0

3:2

r1N IL

r0N
V0

rN
;
rN r0N

VL
rN

RL

Fig. 3.1 Electric circuit with a variable series resistance

Fig. 3.2 Thvenin equivalent


circuit with a variable internal
resistance and voltage source

Ri

IL

VL
OC

VL

RL

3.2 Circuit with a Series Variable Resistance


Fig. 3.3 Two load straight
lines with the parameters r10N,
r20N

57

IL
1

Ri

SC1

IL

IL

r0 N
2

Ri

RL

SC2

r0 N

OC1

VL

V L VL

OC2

VL

IL

Ri r1N
ILSC

r0N rN
;
r0N rN

VLOC
V0



:
Ri
r0N 1 rr1NN r1N

3:3
3:4

1
2
Let the resistance r0N vary from r0N
to r0N
. Then, we get the two load straight
lines in Fig. 3.3 with the following parameters of the Thvenin equivalent circuit

VLOC1 ; VLOC2 ;

R1i ; R2i ;

ILSC1 ; ILSC2 :

Expression (3.1) allows calculating the load current for the given load resistance.
But, the recalculation of all the parameters of the equivalent circuit is necessary that
denes the disadvantage of this known Thvenin equivalent circuit.

3.2.2

Generalized Equivalent Circuit

According to Fig. 3.3, these load straight lines intersect into a point G. Physically, it
means that regime parameters do not depend on the value r0N; that is, the current
across this element is equal to zero at the expense of the load value.
In this case, the point G will be a bunch center of the load straight lines with the
parameter r0N.
Let us dene this bunch center. At once, it is obvious that the current across the
resistance r0N will be equal to zero if the voltage VN = V0. Then, we get a circuit in
Fig. 3.4 for the calculation of the load parameters.

58

3 Generalized Equivalent Circuit of an Active Two-Pole


G

IL

V0 r1N
G

rN

VL

+
-

R L <0

Fig. 3.4 Part of the circuit with zero current across the variable resistance r0N

Then, the load current and voltage


ILG
VLG V0 r1N ILG

V0
;
rN

3:5

r1N rN
V0 [ V0 :
rN

3:6

The load resistance is a negative value


RG
L

VLG
r1N rN :
ILG

3:7

Thus, an equation of straight line, passing through a point I1G ; V1G , has the form
IL ILG

VLG VL
 :
Ri
Ri

3:8

So, the values I1G ; V1G , and Ri are the parameters of the generalized
Thvenin/Helmholtz equivalent generator in Fig. 3.5.
We note that besides a base energy source of one kind (a voltage source VLG )
there is an additional energy source of another kind (a current source ILG ) that it is
possible to consider as a corresponding theorem.
It is natural, when the value ILG 0, we obtain the known Thvenin/Helmholtz
equivalent generator. In this case VLG VLOC . Using the generalized generator, we
must recalculate the internal resistance only. It is the advantage of the offered
generator.
Let us show how the internal resistance Ri and changeable resistance r0N,
respectively influence on properties of the generalized equivalent generator in
Fig. 3.5. The corresponding family of the load straight lines is shown in Fig. 3.6.

3.2 Circuit with a Series Variable Resistance

59

Ri

IL

VL

IL

VL

RL

Fig. 3.5 Generalized Thvenin/Helmholtz equivalent generator

Fig. 3.6 Family of the load


straight lines with the
characteristic values Ri and
r0N

IL

r0 N
RL

Ri =0

r0 N
Ri
0

r 0 N =
0

Ri

VL

VL

r 0N
I

R i =

IL

Further, we use the inverse expression to (3.3); that is,


r0N

rN Ri  r1N
:
rN  Ri  r1N

3:9

The resistances Ri, r0N have the following characteristic values:


RVi 0;

V
r0N


rN r1N
;
rN r1N

3:10

which denes the generalized equivalent generator as an ideal voltage source;


RIi 1;

I
r0N
rN ;

3:11

which denes the generalized equivalent generator as an ideal current source;


R0i RG
L r1N rN ;

0
r0N
1;

3:12

which corresponds to the beam G0 and denes the zero-order source when the
current and voltage of the load are always equal to zero for all the load values.
The generalized equivalent generator that displays the zero-order generator is
presented in Fig. 3.7. VL = 0 because the internal resistance voltage V0i = VG
L.

3 Generalized Equivalent Circuit of an Active Two-Pole

60

IL = 0

Vi

VL

VL = 0

Ri

IL

RL

Fig. 3.7 Zero-order generator

So, a variable element and internal resistance can have these three specied
characteristic values. These values are dened at a qualitative level. This brings up
the problem of determination in the relative or normalized form of the value r0N
regarding of these characteristic values. In this case, it is possibly to dene a kind of
active two-pole as an energy source and to compare the different circuits. Therefore,
the obvious value r0N = 0 is not characteristic one concerning the load.
Let us view the possible load characteristic values. Both the traditional values
RL = 0, RL = , and RG
L will be the characteristic values according to Fig. 3.6.
Physical sense of these values is clear.

3.2.3

Relative Operative Regimes. Recalculation of the Load


Current

Let us consider a common change of load RL and variable resistance r0N of the
circuit in Fig. 3.1. The corresponding load straight lines are shown in Fig. 3.8.
In this gure, we get the two bunches with parameters RL and r0N. Let an initial
1
2
and subsequent value is r0N
. Similarly, an
value of the variable element be r0N
initial value of the load equals R1L and subsequent one is R2L .
Fig. 3.8 Common change of
load RL and variable
resistance r0N

IL R =0
L
1

r 0N

F2

CI

r 0N

F1

BI
B2

RL

RL

C2
C1

R i = RL

B1
0

r 0N
A2

IL

RL=

CV

r0N

A1

r 0N

BV

VL

VL
G

RL

3.2 Circuit with a Series Variable Resistance

61

Let us consider the straight line of the initial load R1L . The three straight lines
I
0
V
with the characteristic values r0N
; r0N
; r0N
and the two lines with the parameters
1
2
1
r0N , r0N intersect this line RL . The points A1, 0, B1, C1, F1 are points of this
intersection. In turn, the points A2, 0, B2, C2, F2 are points of intersection of the line
R2L . Therefore, a projective map (conformity) of one line R1L on the other line R2L
takes place. This conformity is set by a projection center G.
1
. The three straight lines with the
Similarly, we consider the straight line r0N
characteristic values of load RL and two lines with parameters R1L , R2L intersect this
1
line r0N
. The points G, CV, C1, C2, CI are points of this intersection.
2
In turn, the points G, BV, B1, B2, BI are points of intersection of the line r0N
.
1
2
Therefore, a projective map (conformity) of one line r0N
on the other line r0N
takes place. This conformity is set by the projection center 0.
The above conformities of straight line points represent projective geometry
transformations. As shown earlier, it is convenient to use projective geometry for
analysis of circuits with variable elements. The projective transformation is also set
by three pairs of respective points. As pairs of these points, it is convenient to use
the points corresponding to the characteristic values of the load and variable element. The projective transformations preserve a cross ratio of four points; otherwise, a cross ratio is an invariant of these transformations. Further, we will show the
use of such invariants.
Case 1 Denition of the relative operating regime at the load change
Let the series resistance r0N be equal r10N and the load resistance varies from R1L to
R2L . In this case, a point of initial regime C1 C2, as it is shown in Fig. 3.8. If the
2
series resistor is equal to r0N
, a point of initial regime B1 B2. The given points CI,
C2, C1, CV, G correspond to the points BI, B2, B1, BV, G.
Using (2.24), we may constitute the cross ratio for the initial regime in the form
m1L CI C1 CV G BI B1 BV G
0 R1L 1 RG
L

R1L

R1L
:
 RG
L

3:13

The points CI, G (and BI, G) are the base points, and CV (and BV) is a unit one.
Likewise, we have the base values RL 0; RG
L , and a unit RL = .
This determination of relative regime does not depend on r0N. Therefore, we
must use the load voltage for the cross ratio calculation. In this case, the base points
CI, BI give the common base point VL = 0.
Similarly to (2.23), the equation VL(RL) from (3.8) has the fractionally linear
view; it is possible by formalized method to express the cross ratio right now

3 Generalized Equivalent Circuit of an Active Two-Pole

62
Fig. 3.9 Mutual conformity
of the load voltage, resistance,
and cross ratio

CI

C2

VL

C1

CV

C1

C2

VL

21

RL

VL

VL

mL

RL

mL1

m L1

RL

VLC1  0
VLCV  0

VLC1  VLG VLCV  VLG
V B1  0
VLBV  0
VLG B1L

:
VL  VLG VLBV  VLG

m1L 0 VLC1 VLCV VLG


0

VLB1

VLBV

VL

mL1
0

CV

RL

3:14

In this case, we have the base values VL 0; VLG , and a unit value VLCV .
The corresponding values of this cross ratio are shown in Fig. 3.9.
So, it is possible to consider cross ratio (3.13) and (3.14) as a projective coordinate of the initial or running regime points VLC1 , R1L . This coordinate is expressed
by invariant (identical) manner by various regime parameters.
The cross ratio for the subsequent regime
m2L CI C2 CV G BI B2 BV G
0 R2L 1 RG
L

R2L

R2L
:
 RG
L

VLC2  0
VLCV  0
 CV
C2
G
VL  VL VL  VLG
V B2  0
VLBV  0
VLG B2L

:
VL  VLG VLBV  VLG

m2L 0 VLC2 VLCV VLG


0

VLB2

VLBV

3:15

3:16

Similarly to (2.33) and (2.34), the regime change has the view
m21
L CI C2 C1 G BI B2 B1 G
0 R2L R1L RG
L

R2L
R1L

m2L  m1L ;
R2L  RG
R1L  RG
L
L

3:17

3.2 Circuit with a Series Variable Resistance

63

VLC2
VLC1

VLC2  VLG VLC1  VLG
V B2  0
VLB1  0
VLG B2L

:
VL  VLG VLB1  VLG

C2 C1 G
m21
L 0 VL VL VL

VLB2

VLB1

3:18

This change is expressed by invariant manner through various regime parameters. Therefore, usually used regime changes by increments (as formal) are
eliminated.
In turn, the values VLG , RG
L are the scales for normalizing the voltage and
resistance values. Then, expressions (3.17) and (3.18) represent the relative
regimes. It permits to compare or set the regime of different circuits with various
parameters.
Let us remind properties of a cross ratio. If the components VLC1 , VLC2 of
expression (3.18) are interchanged, we get
m12
L

1
:
m21
L

3:19

Also, the group property takes place


2
32 21 1
31 1
m3L m32
L mL mL mL mL mL mL :

3:20

Let us obtain the subsequent voltage value from expression (3.18). Then, we
have
VLC2

VLG VLC1 m21


L
:
G
VLC1 m21
L  1 VL

3:21

The obtained transformation with a parameter m21


L allows realizing the direct
recalculation of load voltage at load change. This expression is especially convenient for the set of the load changes on account of group property (3.20).
Case 2 Denition of the relative operating regime at the change of the series
resistance r0N or internal resistance Ri
The cross ratio m1i of four points, three of these are the characteristic 0, A1, F1 of the
1
line R1L , and the fourth point C1 of the initial regime r0N
or R1i , has the view
m1i 0 C1 A1 F1

C1  0
A1  0

:
C1  F1 A1  F1

3:22

The points 0, F1 are chosen as the base points; that will be explained later. In
turn, A1 is a unit point.

3 Generalized Equivalent Circuit of an Active Two-Pole

64

The cross ratio for the points 0, C2, A2, F2 of the line R2L has the same value
m1i 0 C2 A2 F2

C2  0
A2  0

:
C2  F2 A2  F2

3:23

Cross ratio (3.22) is expressed by voltage components


m1i 0 VLC1 VLA1 VLG

VLC1  0
V A1  0
 A1L
:
C1
G
VL  VL VL  VLG

3:24

Similarly, the cross ratio for the subsequent regime


m2i 0 VLB1 VLA1 VLG

VLB1  0
VLA1  0
:
G  A1
B1
VL  VL VL  VLG

3:25

The distance between the points of the initial and subsequent regimes on the
straight line R1L
2
1
m21
i mi  mi

VLB1  0
VLC1  0
 C1
0 VLB1 VLC1 VLG :
G
B1
VL  VL VL  VLG

3:26

The same distance of the points on the line R2L


B2 C2 G
m21
i 0 VL VL VL

VLB2  0
VLC2  0
:
G  C2
B2
VL  VL VL  VLG

3:27

For a given load, the equation VL(Ri) = VL(r0N) from expression (3.8) has
fractionally linear view; it is possible, by the formalized method, to express cross
ratio (3.24) and (3.26) for the resistance Ri and r0N
m1i 0 VLC1 VLA1 VLG R0i R1i RIi RVi
1
RG
L Ri 1 0

R1i RG
1 RG
R1 RG
L
L
i 1 L;

1
10
Ri  0
Ri

0
1
I
V
m1i 0 VLC1 VLA1 VLG r0N
r0N
r0N
r0N

1
I
V
1 r0N
r0N
r0N

1
I
I
V
r0N
1
r0N
 1 r0N
 r0N


:
I  rV
1  rV
1  rV
r0N
r0N
r0N
0N
0N
0N

B1 C1 G
0 2 1 V
m21
i 0 VL VL VL Ri Ri Ri Ri
2 1
RG
L Ri Ri 0

R2i RG
R1i RG
L
L

;
R2i
R1i

3:28

3:29

3:30

3.2 Circuit with a Series Variable Resistance

65

B1 C1 G
0
2
1
V
m21
i 0 VL VL VL r0N r0N r0N r0N
2
1 V
1 r0N
r0N
r0N

2
1
1
V
r0N
1
r0N
 1 r0N
 r0N


:
2  rV
1  rV
2  rV
r0N
r0N
r0N
0N
0N
0N

3:31

The subsequent voltage value from expression (3.26) is obtained


VLB1

VLG VLC1 m21


i
:
G
VLC1 m21
i  1 VL

3:32

This transformation with a parameter m21


i allows realizing the direct recalculation of load voltage.
Case 3 Denition of the relative operating regime at the common change of the
load RL and resistance r0N
Let a common or compound change of regime be given as C1 C2 B2.
Then, the view of expressions (3.27) and (3.18) shows that it is possible to use
the multiplication of these cross ratios as the compound regime change
21
m21 m21
i mL

VLB2  0
VLC1  0

0 VLB2 VLC1 VLG :
VLB2  VLG VLC1  VLG

3:33

In this resultant expression, intermediate components are reduced at the expense


of the identical base points. Therefore, we obtain the resultant voltage value for the
point B2
VLB2

3.2.4

VLG VLC1 m21


C1
VL m21  1 VLG

3:34

Example

Let us consider a circuit with given values in Fig. 3.10. Dimensions of these values
are not indicated for simplifying.
1
0:5.
Let the initial value of the resistance r0N be equal to r0N
Parameters (3.2)(3.4) of the Thvenin equivalent circuit
VLOC1 V0

rN
5
9:0909;
10
1
5 0:5
rN r0N

R1i r1N
ILSC1

1
r0N
rN
0:5  5
1:4545;
1
1
0:5 5
r0N rN

VLOC1
6:25:
R1i

3 Generalized Equivalent Circuit of an Active Two-Pole

66

r1N IL

r0N
V0

VL

1
rN 5

10

RL

Fig. 3.10 Example of a circuit with given elements


2
Let the subsequent value of the resistance r0N be equal to r0N
1.
The corresponding parameters of the Thvenin equivalent circuit

VLOC2 10

5
8:3333;
51

15
1:8333;
15

R2i 1

ILSC2 4:5454:

The corresponding load straight lines are shown in Fig. 3.11.


The parameters of the known equivalent generator correspond to points CV, CI,
and points BV, BI; that is,
VLCV 9:0909;

ILCI 6:25;

VLBV 8:3333;

ILBI 4:5454:

Bunch center coordinates (3.5), and (3.6)


ILG

V0 10
2;

5
rN

VLG

Fig. 3.11 Example of load


straight lines of a circuit with
the variable series resistance

r1N rN
15
10 12:
V0
5
rN

IL
1

r 0N
6

F2

RL

CI

r 0N

BI

C3

B2

F1
C1
B1

r 0N=
4

A1

r 0N

RL

C2

r 0N

CV
6

BV

VL
10

12

VL

RL

A2

IL

3.2 Circuit with a Series Variable Resistance

67

Ri

VL= 0

VL

12

IL

12

RL

Fig. 3.12 Example of the zero-order generator

Corresponding negative load resistance (3.7)


RG
L

VLG
6:
ILG

Characteristic values (3.10)(3.12) of variable element and internal resistance


RVi 0;
RIi 1;

51
0:8333;
51
rN 5;

V
r0N

I
r0N

R0i RG
L 6;

0
r0N
1:

Figure 3.12 demonstrates the zero-order generator.


Case 1 Recalculation of the load voltage at the load change
1
and the load resistance varies from R1L to
Let the series resistance r0N be equal to r0N
R2L . In this case, the initial regime point C1 C2.
We consider the load voltage of the initial regime VC1
L = 6. Using expression
(3.8) of the generalized equivalent generator, we get the load current

ILC1 ILG

VLG  VLC1
12  6
2:1251:
2
1
1:4545
Ri

Then, the load resistance


R1L

VLC1
6
2:8233:

C1
2:1251
IL

We consider the load voltage of the subsequent regime VC2


L = 4.5. We get the
load current and resistance
ILC2 3:1564;

R2L 1:4256:

3 Generalized Equivalent Circuit of an Active Two-Pole

68

2
If the series resistor is equal to r0N
, the initial regime point B1 B2. Then,

VLB1 5:0524;

VLB2 3:6455:

Initial regime cross ratio (3.13)


m1L CI C1 CV G BI B1 BV G

R1L
2:8233
0:32:

2:8233
6
R1L  RG
L

We check cross ratio (3.14)


VLC1  0
VLCV  0
 CV
C1
G
VL  VL VL  VLG
60
9:0909  0

1  3:1249 0:32;

6  12 9:0909  12
V B1  0
VLBV  0
m1L B1L

VL  VLG VLBV  VLG
5:0524  0
8:3333  0

0:7272  2:2726 0:32:

5:0524  12 8:3333  12

m1L

Subsequent regime cross ratio (3.15)


m2L CI C2 CV G BI B2 BV G

R2L
1:4256
0:192:

G
2
RL  RL 1:4256 6

Regime change (3.17)


m21
L

R2L
R1L

0:192  0:32 0:6:
R2L  RG
R1L  RG
L
L

We may check subsequent value (3.18) of load voltage


VLC2

VLG VLC1 m21


12  6  0:6
43:2
L

4:5:

C1
G
21
6  0:6  1 12
9:6
VL mL  1 VL

We consider the load resistance R3L = 0.7163 that corresponds to the point C3 and
cross ratio
m3L CI C3 CV G

RL
0:7163
0:1066:

0:7163
6
R3L  RG
L

3.2 Circuit with a Series Variable Resistance

69

Regime change (3.20)


m32
L 0:1066  0:192 0:5555:
Using group property (3.20), we may directly calculate the load voltage
VLC3

VLG VLC2 m32


L
C2
VL m32

1
VLG
L

12  4:5  0:5555
30

3:0:
4:5  0:5555  1 12 10

Case 2 Recalculation of the load voltage at the series resistance r0N change
1
2
!r0N
. In this
Let the load resistance be equal to R1L and the series resistance r0N
case, the initial regime point C1 B1.
Regime change (3.31)

2
1
V
m21
i 1 r0N r0N r0N

1
V
r0N
 r0N
0:5 0:8333
V 1 0:8333 0:7272:
2
r0N  r0N

We consider the load voltage of the initial regime VLC1 6. Then, subsequent
voltage value (3.32)
VLB1

VLG VLC1 m21


i
C1
VL m21

1
VLG
i

12  6  0:7272
5:0523:
6  0:7272  1 12

Similarly, let the load resistance be equal to R2L ; the point of the initial regime
C2 B2. We consider the load voltage of the initial regime VLC2 4:5. Then, the
subsequent voltage value
VLB2

VLG VLC2 m21


12  4:5  0:7272
i
3:6453:

C2
G
21
4:5  0:7272  1 12
VL mi  1 VL

Case 3 Recalculation of the load voltage at the common change of the load
RL and resistance r0N
Let the regime change be given as C1 C2 B2.
Common regime change (3.33)
21
m21 m21
i mL 0:7272  0:6 0:4363:

Resultant voltage value (3.34) for the point B2


VLB2

VLG VLC1 m21


12  6  0:4363
3:6453:

C1
G
21
6  0:4363  1 12
VL m  1 VL

3 Generalized Equivalent Circuit of an Active Two-Pole

70

3.3
3.3.1

Circuit with a Shunt Variable Conductivity


Disadvantage of the Known Equivalent Circuit

Let us consider an active two-pole circuit with a base load conductivity YL and
variable auxiliary load or shunt regulating conductivity yN in Fig. 3.13. This circuit
has a practical importance for a current regulation.
At change of the load conductivity from the short circuit SC to open circuit OC
for the specied shunt conductivity yN, a load straight line is given by expression
(3.1)
IL Yi VLOC  Yi VL ILSC  Yi VL ;

3:35

where VLOC is the OC voltage; the internal conductivity Yi and SC current ILSC are the
parameters of the Norton equivalent circuit in Fig. 3.14.
For our two-pole circuit we have
VLOC V0
Yi

y0N
;
yN y0N

y0N yN
y1N ;
y0N yN y1N

ILSC Yi VLOC V0

y0N y1N
:
y1N y0N yN

3:37
3:38

y1N IL

y0N
V0

3:36

VL
yN

YL

Fig. 3.13 Electric circuit with a variable shunt conductivity

IL

VL
SC

IL

Yi
YL

Fig. 3.14 Norton equivalent circuit with the variable internal conductivity and current source

3.3 Circuit with a Shunt Variable Conductivity

71

Fig. 3.15 Two load straight


lines with parameters y1N, y2N

IL

IL

YL

SC1

IL
SC2

IL

OC2

VL

OC1

VL

VL

VL

Y i y2
N

Y i y1
N

Let the conductivity yN varies from y1N to y2N . Then, we get the two load straight
lines in Fig. 3.15 with the following parameter of the Norton equivalent circuit
VLOC1 ; VLOC2 ;

Yi1 ; Yi2 ;

ILSC1 ; ILSC2 :

Expression (3.35) allows calculating the load current for the given load resistance. But, the recalculation of all parameters of the equivalent circuit is necessary
that denes the disadvantage of this known Norton equivalent circuit.

3.3.2

Generalized Equivalent Circuit

According to Fig. 3.15, these load straight lines intersect into a point G. Physically,
it means that regime parameters do not depend on the value yN; that is, the current
through this element is equal to zero at the expense of the load value.
In this case, the point G will be a bunch center of load straight lines with the
parameter yN.
Let us dene this bunch center. At once, it is visible that the current across yN
will be equal to zero if the voltage VN = 0. Then, we get a circuit in Fig. 3.16 for the
calculation of the load parameters.

y1N I L

y0N
V0

VN=0

VL

Y L <0

Fig. 3.16 Circuit with the zero voltage of the variable conductance yN

3 Generalized Equivalent Circuit of an Active Two-Pole

72

Then, the current across the conductivity y0N is


ILG y0N V0 :

3:39

The voltage through the conductivity y1N is


V1N

ILG
:
y1N

On the other hand,


VLG V1N VN 0:
Therefore
VLG V1N

y0N
V0 :
y1N

3:40

In turn, the load conductivity is a negative value


YLG

ILG
y1N :
VLG

3:41

Thus, an equation of a straight line, passing through a point I1G ; V1G , has the form
IL  ILG Yi VLG VL :

3:42

So, the values ILG ; VLG , and Yi are the parameters of the generalized
Norton/Mayer equivalent generator in Fig. 3.17.
We note that besides a base energy source of one kind (a current source IG
L)
there is an additional energy source of another kind (a voltage source VG
L ) that it is
possible to consider as a corresponding theorem.
It is natural, when the value VG
L = 0, we obtain the known Norton/Mayer
SC
equivalent generator. In this case IG
L = IL . Using the generalized generator, we must

VL

IL
VL

IL

Yi

Fig. 3.17 Generalized Norton/Mayer equivalent generator

YL

3.3 Circuit with a Shunt Variable Conductivity

73

Fig. 3.18 Family of the load


straight lines with the
characteristic values Yi and yN

IL

IL

yN
I

Y i=0
YL1

VL

0
0

y N =

Yi

yN

Yi

VL

yN
Y i =
V

recalculate the internal conductivity value only. It is the advantage of the offered
generator.
Let us demonstrate how the internal conductivity Yi and respectively the
changeable conductivity yN influence on properties of the generalized equivalent
generator in Fig. 3.17. The corresponding family of the load straight lines is shown
in Fig. 3.18.
Further, we use expression (3.37).
The conductivities Yi, yN have the following characteristic values:
YiI 0;

yIN y0N ;

3:43

which denes the generalized equivalent generator as an ideal current source;


YiV 1;

yVN y0N y1N ;

3:44

which denes the generalized equivalent generator as an ideal voltage source;


Yi0 y1N YLG ;

y0N 1;

3:45

which corresponds to the beam G0 and denes the zero-order source when the
current and voltage of the load are always equal to zero for all load values.
The generalized equivalent generator that exhibits the zero-order generator is
presented in Fig. 3.19. The load voltage VL = 0 because V0i = VG
L.
So, a variable element and internal conductivity can have these three specied
characteristic values. This brings up the problem of determination in the relative or
normalized form of the conductivity value yN regarding of these characteristic
values. Therefore, the obvious value yN = 0 is not characteristic ones concerning the
load.

3 Generalized Equivalent Circuit of an Active Two-Pole

74

IL= 0

VL
0

IL

Yi

VL= 0
0

Vi

YL

Fig. 3.19 Zero-order generator

Let us view the load characteristic values. Both the traditional values YL = 0,
YL = , and YLG will be the characteristic values according to Fig. 3.18.

3.3.3

Relative Operative Regimes. Recalculation of the Load


Current

Let us consider a common change of load YL and variable shunt conductivity yN of


the circuit in Fig. 3.13. The corresponding load straight lines are shown in Fig. 3.20.
In this gure, we get the two bunches with parameters YL and yN. Let an initial
value of the variable element be y1N and subsequent value is y2N . Similarly, an initial
value of the load equals YL1 and subsequent one is YL2 .
Fig. 3.20 Common change
of load YL and variable
conductivity yN

IL
G

YL

YL=

IL

F2

YL

F1

YL
I

yN

G
CI
C2
BI
G

VL

B2

C1
B1

BV

A1
V

yN
A2

yN
2

yN

YL=0

CV

yN

VL

3.3 Circuit with a Shunt Variable Conductivity

75

Case 1 Denition of the relative operating regime at the load change


The cross ratio m1L of the four point, three of these are the characteristic points CV,
CI, G of the line y1N , and the fourth C1 is the point of the initial regime YL1, has the
view
C1  CV CI  CV

C1  G
CI  G
1
Y
0 YL1 1 YLG 1 L G :
YL  YL

m1L CV C1 CI G

3:46

The cross ratio for the points BV ; B1 ; BI ; G of the line y2N has the same value
m1L BV B1 BI G

B1  BV BI  BV

:
B1  G
BI  G

3:47

The points CV, G (and BV, G) are the base points, and CI (and BI) is a unit one.
Correspondingly, we have the base values yL = 0, yG
L , and a unit yL = . This
determination of relative regime does not depend on YN. Therefore, we must use the
load current for the cross ratio calculation. In this case, the base points CV, BV give
the common base point IL = 0.
Then, the cross ratio is expressed by current components
m1L 0 ILC1 ILCI ILG

ILC1  0
ILCI  0

:
ILC1  ILG ILCI  ILG

3:48

The corresponding values of this cross ratio are shown in Fig. 3.21.
Similarly, the subsequent regime cross ratio
m2L 0 ILC2 ILCI ILG

Fig. 3.21 Mutual conformity


of the load current,
conductivity, and cross ratio

ILC2  0
ILCI  0

:
ILC2  ILG ILCI  ILG

3:49

CI

C2

C1

CI

C2

C1

IL

IL

IL

mL

YL

IL

IL

mL

YL

21

mL

mL

YL

CV

YL

76

3 Generalized Equivalent Circuit of an Active Two-Pole

The regime change has the view


m21
L CV C2 C1 G BV B2 B1 G
0 YL2 YL1 YLG

YL2
YL1  0

:
YL2  YLG YL1  YLG

C2 C1 G
m21
L 0 IL IL IL

ILC2  0
ILC1  0

:
ILC2  ILG ILC1  ILG

3:50

3:51

This change is expressed by invariant manner through various regime parameters. Therefore, usually used regime changes by increments (as formal), are
eliminated.
In turn, the values ILG , YLG are the scales for normalizing the current and conductivity values. Then, expressions (3.50) and (3.51) represent the relative regimes.
It permits to compare or set the regime of the different circuits with various
parameters.
Also, the group property takes place
2
32 21 1
31 1
m3L m32
L mL mL mL mL mL mL :

3:52

Let us obtain the subsequent current value from expression (3.51). Then, we
have
ILC2

ILG ILC1 m21


L
C1
G
IL m21
L  1 IL

3:53

The obtained transformation with a parameter m21


L allows realizing the direct
recalculation of a load current at a load change. This expression is especially
convenient for the set of the load changes on account of group property (3.52).
Case 2 Denition of the relative operating regime at the change of the shunt
conductivity yN or internal conductivity Yi
The cross ratio m1i of the four points, three of these are the characteristic 0, A1, F1 of
the line YL1 , and the fourth point C1 of the initial regime y1N or Yi1 , has the view
m1i 0 C1 A1 F1

C1  0
A1  0

:
C1  F1 A1  F1

3:54

The points 0, F1 are chosen as the base points; that will be explained later. In
turn, the point A1 is a unit point.
The cross ratio for the points 0; C2 ; A2 ; F2 of the line YL2 has the same value
m1i 0 C2 A2 F2

C2  0
A2  0

:
C2  F2 A2  F2

3:55

3.3 Circuit with a Shunt Variable Conductivity

77

Cross ratio (3.54) is expressed by current components


m1i 0 ILC1 ILA1 ILG

ILC1  0
ILA1  0

:
ILC1  ILG ILA1  ILG

3:56

Similarly, the cross ratio of the subsequent regime


m2i 0 ILB1 ILA1 ILG

ILB1  0
ILA1  0
:
G  A1
B1
IL  IL IL  ILG

3:57

The distance between the points of the initial and subsequent regimes on the
straight line YL1
2
1
m21
i mi  mi

ILB1  0
ILC1  0
0 ILB1 ILC1 ILG :
G  C1
B1
IL  IL IL  ILG

3:58

The same distance of the points on the line YL2


B2 C2 G
m21
i 0 IL IL IL

ILB2  0
ILC2  0

:
ILB2  ILG ILC2  ILG

3:59

For a given load, the expression IL = (Yi) = IL = (yN) has fractionally linear view;
it is possible, by the formalized method, to express cross ratio (3.56) and (3.58) for
the conductivity Yi and yN
m1i 0 ILC1 ILA1 ILG Yi0 Yi1 YiV YiI
YLG Yi1 1 0

Yi1 YLG 1 YLG Yi1 YLG

:

10
Yi1  0
Yi1

m1i 0 ILC1 ILA1 ILG y0N y1N yVN yIN


1 y1N yVN yIN

y1N  1 yVN  1 yVN  yIN




:
y1N  yIN yVN  yIN y1N  yIN

B1 C1 G
2 1 I
m21
i 0 I1 I1 I1 0 Yi Yi Yi

YLG Yi2 Yi1 0

Yi2 YLG Yi1 YLG


 1
:
Yi2  0
Yi  0

B1 C1 G
0 2 1 I
m21
i 0 IL IL IL yN yN yN yN

1 y2N y1N yIN

y2N  1 y1N  1 y1N  yIN




:
y2N  yIN y1N  yIN y2N  yIN

3:60

3:61

3:62

3:63

3 Generalized Equivalent Circuit of an Active Two-Pole

78

The subsequent current value from expression (3.58) is obtained


I1B1

ILG ILC1 m21


i
:
G
I1C1 m21
i  1 IL

3:64

This transformation with a parameter m21


i allows realizing the direct recalculation of load current.
Case 3 Denition of the relative operating regime at the common change of the
load YL and conductivity yN
Let a common change of regime be given as C1 C2 B2.
Then, the view of expressions (3.58) and (3.49) shows that it is possible to use
the multiplication of these cross ratios as the common regime change
21
m21 m21
i mL

ILB2  0
ILC1  0
 C1
0 ILB2 ILC1 ILG :
G
B2
IL  IL IL  ILG

3:65

In this resultant expression, intermediate components are reduced at the expense


of the identical base points. Therefore, we obtain the resultant current value for the
point B2
ILB2

3.3.4

ILG ILC1 m21


:
ILC1 m21  1 ILG

3:66

Example

Let us consider a circuit with given values in Fig. 3.22.


Let the initial value of the conductivity yN be equal to y1N = 0.1. Parameters
(3.36)(3.38) of the Norton equivalent circuit
VLOC1 V0
Yi1
ILSC1

y0N
1
yN y0N
y1N y0N

10

1
9:0909;
0:1 1
0:1 1

2 0:7096;
y1N
0:1 1 2
y1N y0N y1N
Yi1 VLOC1 0:7096  9:0909 6:4516:

y1N IL

y0N
V0
10

1 y
N

Fig. 3.22 Example of a circuit with given elements

VL
YL

3.3 Circuit with a Shunt Variable Conductivity


Fig. 3.23 Example of load
straight lines of a circuit with
the variable shunt
conductivity

79
G

YL

IL

IL

F2

10

6
4
2

C2

BI

C1
B2
B1
4

CV

BV

0
4

YL

CI

VL

yN

YL

F1

6
2

yN

10

VL

yN

A1

yN

A2

y N =

Let the subsequent value of the conductivity yN be equal to y2N = 1.6987.


The corresponding parameters of the Norton equivalent circuit
1
3:7055;
1:6987 1
1:6987 1
Yi2
2 1:1486;
1:6987 1 2

VLOC2 10

ILSC2 4:2561:

The corresponding load straight lines are shown in Fig. 3.23.


The parameters of the known equivalent generator correspond to points CV, CI,
and points BV, BI; that is,
VLCV 9:0909;

ILCI 6:4516;

VLBV 3:7055;

ILBI 4:2561:

Bunch center coordinates (3.39) and (3.40)


ILG y0N V0 1  10 10;

VLG

y0N
1
V0 10 5:
2
y1N

3 Generalized Equivalent Circuit of an Active Two-Pole

80

VL
0

Yi

IL

VL=0

5
-

YL

Fig. 3.24 Example of the zero-order generator

Corresponding negative load conductivity (3.41)


YLG

ILG
2:
VLG

Characteristic values (3.43)(3.45) of the variable element and internal


conductivity
YiI 0;

yIN y0N 1;

YiV 1;

yVN y0N y1N 3;

Yi0 YLG 2;

y0N 1:

Figure 3.24 demonstrates the zero-order generator.


Case 1 Recalculation of the load current at the load change
Let the shunt conductivity yN be equal to y1N and the load conductivity varies from
Y1L to Y2L. In this case, the initial regime point C1 C2.
We consider the load current of the initial regime IC1
L = 3.5. Using expression
(3.42) of the generalized equivalent generator, we get the load voltage
VLC1 VLG

ILG  ILC1
10  3:5
4:160:
5
1
0:7096
Yi

Then, the load conductivity is


YL1

ILC1
3:5
0:8413:

VLC1 4:16

We consider the load current of the subsequent regime IC2


L = 4.5. We get the load
voltage and conductivity
VLC2 2:75;

YL2 1:635:

3.3 Circuit with a Shunt Variable Conductivity

81

If the shunt conductivity is equal to y2N = 1.6987, the initial regime point
B1 B2. Then, we obtain
ILB1 1:8;

ILB2 2:5:

Cross ratio (3.46) for the initial regime


m1L CV C1 CI G BV B1 BI G

YL1

YL1
0:8413
0:296:

 YLG 0:8413 2

We check cross ratio (3.48)


ILC1  0
ILCI  0
 CI
C1
G
IL  IL IL  ILG
3:5  0
6:4516  0


3:5  10 6:4516  10
0:5384  1:8181 0:296:

m1L

Subsequent regime cross ratio (3.49)


m2L

ILC2  0
ILCI  0
 CI
C2
G
IL  IL IL  ILG

0:8181  1:8181 0:45:


Regime change (3.50)
m21
L

YL2

YL2
YL1
0:45  0:296 1:5202:
G 1
 YL YL  YLG

We may check subsequent value (3.53) of load current


ILC2

ILG ILC1 m21


L
C1
IL m21

1
ILG
L

10  3:5  1:5207
4:5:
3:5  0:5207 10

Case 2 Recalculation of the load current at the shunt conductivity yN change


Let the load conductivity be equal to Y1L and the shunt conductivity varies from y1N
to y2N. In this case, the initial regime point C1 B1.
Regime change (3.63)
2 1 I
m21
i 1 yN yN yN

y1N  yIN
0:1 1
0:407:

y2N  yIN 1:6987 1

3 Generalized Equivalent Circuit of an Active Two-Pole

82

We consider the load current of the initial regime IC1


L = 3.5. Then, subsequent
current (3.64)
ILB1

ILG ILC1 m21


10  3:5  0:407
i
1:8:

G
3:5
 0:407  1 10
ILC1 m21

1

I
L
i

Case 3 Recalculation of the load current at the common change of the load
YL and conductivity yN
Let the common change of regime be given as C1 C2 B2.
Common regime change (3.65)
21
m21 m21
i mL 0:407  1:5207 0:6187:

Resultant current (3.66) for the point B2


ILB2

3.4

ILG ILC1 m21


10  3:5  0:6187
2:5:

ILC1 m21  1 ILG 3:5  0:6187  1 10

General Case of an Active Two-Pole with a Variable


Conductivity

Let us consider an active two-pole A with a load conductivity YL1 and variable
auxiliary conductivity YL2 in Fig. 3.25. For convenience of a mathematical
description, the variable element YL2 is taken out from the two-pole contour.
This circuit (as a two-port network) is described by the following system of the
Y parameter equations
"

I1
I2

Y11

Y12

 " # " SC;SC #


I1
V1
Y12

SC;SC :
Y22
V2
I2

YL2

I1
A

V1
YL1

Fig. 3.25 Active two-pole with a load conductivity YL1 and variable element YL2

3:67

3.4 General Case of an Active Two-Pole with a Variable Conductivity

83

The short circuit SC currents of all the loads


I1SC;SC Y10 V0 ;

I2SC;SC Y20 V0 ;

where V0 is a resultant value of sources entering into this active two-pole.

3.4.1

Known Equivalent Generator

Taking into account the current


I2 YL2 V2 ;

3:68

we obtain an expression of a load straight line



I1  Y11 




2
Y12
Y12 Y20
V1 Y10
V0 :
YL2 Y22
YL2 Y22

3:69

The load straight line with the parameter YL2 is shown in Fig. 3.26.
Let us consider the short circuit current ISC
1 . In this case V1 = 0 and
I1SC



Y12 Y20
Y10
V0 :
YL2 Y22

3:70

Similarly, we consider the open circuit voltage VOC


1 . In this case I1 = 0 and
V1OC

Fig. 3.26 Load straight lines


with the parameter YL2

1 SC
I ;
Yi 1

3:71

I1
YL1=

SC

YL1

I1

I1
YL1 = 0
0

V1

OC

V1

YL2

3 Generalized Equivalent Circuit of an Active Two-Pole

84

where the value


Yi Y11 

2
Y12
YL2 Y11 DY
1

Ri
YL2 Y22
YL2 Y22

3:72

is the internal conductivity (internal resistance Ri) of the circuit relatively to the load
YL1; DY is the determinate of the matrix Y parameters.
Taking into account the entered parameters, Eq. (3.69) becomes as
I1 Yi V1 I1SC :

3:73

This expression we represent as


I1  I1SC Yi V1 :

3:74

Thus, we obtain an equation of a straight line passing through a point ISC


1 . In
turn, conductivity Yi denes a slope angle of this line.
So, the values ISC
1 , Yi are the parameters of the Norton equivalent circuit.
By analogy to (3.74) and taking into account (3.71), we obtain an equation of a
straight line passing through a point VOC
1
I1 V1OC  V1 Yi :

3:75

So, the values VOC


1 , Yi = 1/Ri are the parameters of known Thvenin equivalent
OC
circuit. We note that the parameters ISC
depend on the changeable conduc1 , V1
tivity YL2.

3.4.2

Generalized Equivalent Circuit

Let us study features of load straight line (3.69). This expression represents a bunch
G
of the load straight lines with the parameter YL2. To nd the coordinates VG
1 , I1 of
the bunch center G of these lines, it is convenient to use the extreme parameter
values; that is, YL2 = 0, YL2 = . These lines are shown in Fig. 3.27.
In this case, expression (3.69) gives the following system of equations
(




Y2
I1 0  Y11  Y1222 V1 Y10
I1 1 Y11 V1 Y10 V0 :

At the point of intersection we have that


I1G I1 0 I1 1:

Y12 Y20
Y22


V0

3:76

3.4 General Case of an Active Two-Pole with a Variable Conductivity


Fig. 3.27 Bunch of load
straight lines with the
parameters YL2

85

I1
G

Y L1

YL1

I1

SC

I1

I1
OC

V1
G

V1

V1

YL2=

YL2

V1
YL2= 0

The solving of the system gives the following values


Y20
V0 ;
Y12


Y11 Y20
I1G Y10
V0 :
Y12
V1G 

3:77
3:78

G
The obtained values VG
1 , I1 , and the internal conductivity Yi allow to represent
Eq. (3.69) or (3.74) in the other kind

I1  I1G Yi V1 V1G :

3:79

G
Thus, we obtain an equation of a straight line passing through a point IG
1 , V1 . So,
G
G
the values I1 , V1 , and Yi are the parameters of the generalized Norton/Mayer
equivalent circuit in Fig. 3.17 of Sect. 3.2. In turn, this equation corresponds to
expression (3.42).
G
Let us note the bunch center G corresponds to such a voltage VG
1 and current I1
G
of the load YL1 = YL1 when the current of the element YL2 is equal to zero.
According to this condition, from (3.67), it is also possible to nd values (3.77) and
G
(3.78) of VG
1 , I1 . Then, the corresponding negative load conductivity

G
YL1

I1G
Y11 Y20 Y10 Y21

:
Y20
V1G

3:80

In the above case, the bunch center G is in the second quadrant of the coordinate
G
system and so VG
1 < 0, I1 > 0. It is natural to consider a case when the bunch center
G is in the fourth quadrant shown in Fig. 3.28.

3 Generalized Equivalent Circuit of an Active Two-Pole

86
Fig. 3.28 Bunch center is in
the fourth quadrant

I1

YL1

YL2 =
YL2 =0

V1

I1

V1
G

R L1

To do this, we may consider expression (3.80) as follows


G
YL1

I1G
:
V1G

3:81

Then, expressions (3.77) and (3.78) become as


V1G

Y20
V0 :
Y12



Y11 Y20
I1G  Y10
V0 :
Y12

3:82
3:83

Thus, Eq. (3.79) has the form


I1 I1G 

1
V1  V1G :
Ri

3:84

G
So, the values IG
1 , V1 , and Ri are the parameters of the generalized
Thvenin/Helmholtz equivalent generator in Fig. 3.5 of Sect. 3.1. In turn, Eq. (3.84)
corresponds to expression (3.8).
The position of center G (in the second or fourth quadrant) is dened by the kind
or type of an active two-pole as an energy source. If an active two-pole shows more
properties of current source, the case of Fig. 3.27 takes place. If it shows more
properties of voltage source, we have the case of Fig. 3.28.
Let us demonstrate how the internal conductivity Yi and respectively the conductivity YL2 influence on properties of the generalized equivalent generator.

3.4 General Case of an Active Two-Pole with a Variable Conductivity

87

Further, we use the inverse expression to (3.72)


YL2

Y22 Yi  DY
:
Y11  Yi

3:85

The conductivities have the following characteristic values:


YiI 0;

I
YL2


DY
;
Y11

3:86

which denes the generalized equivalent generator as an ideal current source;


YiV 1;

V
YL2
Y22 ;

3:87

which denes the generalized equivalent generator as an ideal voltage source;


I1G
Y11 Y20 Y10 Y21
G
YL1

;
G
Y20
V1
Y20 Y12
Y22 
;
Y10

Yi0 
0
YL2

3:88

which denes a zero-order source.

3.4.3

Example of a Circuit. Recalculation of the Load


Current

Let us consider the electric circuit with the base load conductivity YL1 and auxiliary
load conductivity YL2 in Fig. 3.29.
This circuit may be considered as an active two-port A network relatively to the
specied loads in Fig. 3.30.
Taking into account the specied directions of currents, this network is described
by (3.67)
2
3
" # 
 " #
I1SC;SC
I1
V1
Y11 Y12
5

4

Y12 Y22
I2
V2
I2SC;SC

 " # " #
V1
3
1:2
0:2


0:2 0:95
V2
2

3 Generalized Equivalent Circuit of an Active Two-Pole

88
Fig. 3.29 Active two-pole
with the load conductivity YL1
and variable element YL2

I1

y1N

1.25

y0N
V0
5

Fig. 3.30 Active two-port


A network with the specied
loads

y2N

2.5

I0

yN

0.833

0.625

V1

y1
0.25

I2

y2
0.25

YL1
V2
YL2

V1
A

YL1
V2
YL2

In turn, Y parameters are


y21N
1:5625
1:2;
0:25 1:25 
5:2083
yR
yR y0N yN y1N y2N 2:5 0:625 1:25 0:8333 5:2083;
y
1:25
0:2;
Y12 y2N 1N 0:8333
5:2083
yR
y2
0:6944
0:95:
Y22 y2 y2N  2N 0:25 0:8333 
5:2083
yR

Y11 y1 y1N 

The short circuit SC currents


y1N
1:25
5 0:6  5 3;
V0 2:5
5:2083
yR
y2N
0:8333
5 0:4  5 2:
Y20 V0 y0N
V0 2:5
5:2083
yR

I1SC;SC Y10 V0 y0N


I2SC;SC

Let the initial value of the conductivity YL2 be equal to Y1L2 = 0.5. Parameters
(3.70)(3.72) of the Norton equivalent circuit

3.4 General Case of an Active Two-Pole with a Variable Conductivity


I1SC1

I1CI

89




Y12 Y20
0:2  0:4
Y10 1
 5 3:2758:
V0 0:6
0:5 0:95
YL2 Y22

1
YL2
Y11 DY 0:5  1:2 1:1
1:1724;

1
0:5 0:95
YL2 Y22
1
V1CV 1 I1SC1 2:7941:
Yi

Yi1
V1OC1

Let the subsequent value of the conductivity YL2 be equal to Y2L2 = 2. The
corresponding parameters of the Norton equivalent circuit
I1SC2 I1BI 3:1355;

Yi2 1:1864;

V1OC2 V1BV 2:6428:

The bunch of the load straight lines is shown in Fig. 3.31.


Bunch center coordinates (3.77) and (3.78)
Y20
V1G 
V0 2  5 10;
Y


 12

Y11 Y20
1:2  0:4
G
 5 15:
V0 0:6
I1 Y10
0:2
Y12

Fig. 3.31 Example of load


straight lines of a circuit with
the variable conductivity YL2

Y L1
G

I1
G

I1

YL1=

F2

Y L2

15
2

Y L1=1
CI
3.2758
3.135

BI
C2

1.5073
1.4341
0.979

B2

V1

C1

B1

Y L1=0.5

BV

Y L1=0

CV

V1

10

A1

Y L2
0

Y L2
V

Y L2

Y L2

3 Generalized Equivalent Circuit of an Active Two-Pole

90

The corresponding negative load conductivity


G
YL1

I1G
Y Y20 Y10 Y21
 11
y1N y1 1:5:
Y20
I1G

Characteristic values (3.86)(3.88)


YiI 0;

DY
1:1
0:9166;

1:2
Y11
V
YL2
Y22 0:95;
Y Y20
0
1:5; YL2
Y22  12
y2N y2 1:0833:
Y10

I
YL2


YiV 1;
G
Yi0 YL1

So, the case of the generalized Norton/Mayer equivalent generator takes place.
Therefore, we will use the corresponding relationships of Sect. 3.2.
Case 1 Recalculation of the load current at the load change
Let the conductivity YL2 be equal to Y1L2 and load conductivity varies from Y1L1 = 0.5
to Y2L1 = 1. In this case, the initial regime point C1 C2; the load current of the
initial regime IC1
1 = 0.979.
Cross ratio (3.46) for the initial regime
m1L CV C1 CI G

1
YL1
0:5
0:25:

1  YG
0:5
1:5
YL1
L

Subsequent regime cross ratio (3.49)


m2L

2
YL1
1
0:4:

2  YG
1

1:5
YL1
L

Regime change (3.50)


m21
L 0:4  0:25 1:6:
Subsequent value (3.53) of load current
I1C2

ILG I1C1 m21


L
C1
I1 m21

1
ILG
L

15  0:979  1:6
1:5073:
0:979  0:6 15

Case 2 Recalculation of the load current at the conductivity YL2 change


Let the load conductivity be equal to Y2L1 = 1 and conductivity YL2 varies from
Y1L2 = 0.5 to Y2L2 = 2. In this case, the initial regime point C2 B2.

3.4 General Case of an Active Two-Pole with a Variable Conductivity

91

Cross ratio (3.55) and (3.61) for the initial regime


0
1
V
I
m1i 0 C2 A2 F2 YL2
YL2
YL2
YL2

1
0
0
YL2
 YL2
Y V  YL2
 L2
V
I
1
I
YL2  YL2 YL2  YL2

0:5 1:0833 0:95 1:0833



1:1176  4 0:2794:
0:5 0:9166 0:95 0:9166

Subsequent regime cross ratio


0
2
V
I
m2i 0 B2 A2 F2 YL2
YL2
YL2
YL2

2
0
V
0
YL2
 YL2
YL2
 YL2

V
I
2  YI
YL2
 YL2
YL2
L2

2 1:0833
 4 1:0571  4 0:2642:
2 0:9166

The regime change


2
1
m21
i mi  mi 0:2642  0:2794 0:9459:

Then, subsequent value (3.64) of current for the point B2


I1B2

I1G I1C2 m21


15  1:5073  0:9459
i
1:434:

C2
G
21
1:5073  0:9459  1 15
I1 mi  1 I1

Case 3 Recalculation of the load current at the common change of the load
YL1 and conductivity YL2
Let the common change of regime be given as C1 C2 B2.
The common regime change
21
m21 m21
i mL 0:9459  1:6 1:5134:

Resultant current (3.66) for the point B2


I1B2

3.5

I1G I1C1 m21


15  0:979  1:5134
1:434:

I1C1 m21  1 I1G 0:979  1:5134  1 15

Stabilization of the Load Current

The influence of the conductivity YL2 on properties of the generalized equivalent


generator can be used in practice, for example, for a parametric stabilization of load
current. In this case, the conductivity value YL2 is determined by expression (3.86).
Figure 3.32 schematically shows a structure of the appliance designed to stabilization of the adjustable load current [13]. Power supply voltage source 1 are
connected to the input of line 2 with longitudinal 3 and lateral 4 resistances (loss
resistances). The load 5 is connected to the output of line 2. Negative resistance 6

92

3 Generalized Equivalent Circuit of an Active Two-Pole

Fig. 3.32 Appliance for


stabilization of the load
current

2
1

3
4

V0

VOUT

12

I2

11

V1

V2
10
8

through a match circuit 7 (general case) and a bias voltage source 8 are connected to
one of the lateral resistance 4.
In turn, the negative resistance 6 is determined by a negative-impedance converter with an operation amplier 9. Resistor 10 is an initial positive resistance,
corresponding to the negative resistance 6; resistances 11, 12 are a feedback circuit.
The appliance in Fig. 3.32 work as follows. Let the bias voltage 8 be equal to
zero at rst. The power supply voltage source 1 sets the load current I1 and voltage
V2. The operation amplier 9 gains this voltage V2; the output voltage VOUT
species the current I2. The voltage for the inverse input of the amplier 9 is equal
to V2 and the same current, as I2, flows across the resistor 10. Therefore, the
operation amplier 9 with the resistor 10 is an energy source. This energy source, as
the negative resistance 6, sets the voltage V2, current I2; the ratio V2/I2 is equal to
the resistance 10 value. If the resistance 10 is chosen by necessary way, the load
current will be constant. In turn, the voltage V0 determines the value of the load
current.
For convenience of use of expression (3.86), we may consider a three-port in
Fig. 3.33, which is equivalent to the structure in Fig. 3.32 and the circuit in
Fig. 3.29. In particular, conductivities y0N, y1N correspond to the longitudinal
resistances; conductivities yN, y1 correspond to the lateral resistances of the line.
Also, we have the load conductivity YL1, negative conductivity YL2, and bias
voltage VBS.
For example, let us consider a circuit of ORCAD model in Fig. 3.34. The
negative resistance, by expression (3.86), is equal to 6.63488 k.
The results of modeling for various values of the load resistance RL1 and voltages V0, VBS are given in Tables 3.1 and 3.2.

3.5 Stabilization of the Load Current

93

y0N

I1

y1N

V1

V0
yN

y1
YL1

V2
y2N

I2

YL2

y2
VBS

Fig. 3.33 Equivalent circuit to the structure of stabilization of the load current

I
R1

R6

V
V

8k

6.35V

8k

I0

R3

R7

20k

40k

1.5k

R
V

R11

I
8

V+
+

R2

V2
6.63488k

R12

V-

2k

.
4

L1

20k

OUT

R4

R10

100k

LM358

30V

20k
2V

BS

Fig. 3.34 ORCAD model of stabilization of the load current

We consider Table 3.1. It is visible, that the output voltage VOUT of the operational amplier reaches the maximum value at the load resistance of 3.0 k.
Also, we note that the voltage source 1 does not give but consumes energy in the
given load values because the negative current I0.

94

3 Generalized Equivalent Circuit of an Active Two-Pole

Table 3.1 V0 = 3 V,
VBS = 0 V

RL1 (k)
I1 (mA)
I0 (mA)
V2 (V)
I2 (mA)
VOUT (V)

0.001
0.451
0.076
5.022
0.757
20.17

1.5
0.4508
0.177
6.15
0.9273
24.7

3.0
0.4506
0.2784
7.276
1.097
29.21

Table 3.2 V0 = 6.35 V,


VBS = 2 V

RL1 (k)
I1 (mA)
I0 (mA)
V2 (V)
I2 (mA)
VOUT (V)

0.001
0.4502
0.345
4.175
0.328
10.75

1.5
0.450
0.242
5.3
0.497
15.26

6.1
0.4496
0.0067
8.749
1.017
29.1

Let us consider Table 3.2. The output voltage VOUT of the operational amplier
reaches the maximum value at the load resistance of 6.1 k. It is the advantage of
utilization of a bias voltage source. Also, the voltage source 1 gives energy practically for all load values. Therefore, the use of a bias voltage source improves an
energy effectiveness of power supply voltage source.

References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Chatzarakis, G.E.: A new method for nding the Thvenin and Norton equivalent circuits.
Engineering 2(05), 328336 (2010)
3. Gluskin, E., Patlakh, A.: An ideal independent source as an equivalent 1-port. arXiv preprint
arXiv:1108.4592 (2011)
4. Hashemian, R.: Hybrid equivalent circuit, and alternative to Thvenin and Norton equivalents,
its properties and application. In: Midwest Symposium on Circuits and Systems, MWSCAS
(2009)
5. Hosoya, M.: Derivation of the equivalent circuit of a multi-terminal network given by
generalization of Helmholtz-Thevenins theorem. Bull. Coll. Sci. Univ. Ryukyus. 84, 13
(2007)
6. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
7. Johnson, D.: Origins of the equivalent circuit concept: the voltage-source equivalent. Proc.
IEEE 91(4), 636640 (2003)
8. Karris, S.T.: Circuit Analysis I with MATLAB Computing and Simulink/SimPower Systems
Modeling. Orchard Publications, Fremont (2004)
9. Penin, A.: Utilization of the projective coordinates in the linear electrical network with the
variable regimes. Buletinul Academiei de Stiinte a Republicii Moldova, Fizica si tehnica. 2,
6471 (1992)

References

95

10. Penin, A.: Characteristics of modied equivalent generator of active two-terminal network
with variable resistor. Electrichestvo 4, 5559 (1995)
11. Penin, A.: Linear-fractional relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
12. Penin, A.: Generalized Thvenin/Helmholtz and Norton/Mayer theorems of electric circuits
with variable resistances. WSEAS Trans. Circ. Syst. 13, 104116 (2014). http://www.wseas.
org/cms.action?id=7648. Accessed 30 Nov 2014
13. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer,
Dordrecht (2014)
14. Vandewalle, J.: Shortcuts in circuits and systems education with a case study of the
Thvenin/Helmholtz and Norton/Mayer equivalents. In: International Symposium on the
Circuits and Systems (ISCAS), IEEE (2012)

Chapter 4

Two-Port Circuits

4.1
4.1.1

Input-Output Conformity of Two-Ports


as Afne Transformations
Conformity of a Two-Port

Let us consider a two-port TP1 circuit in Fig. 4.1. A variable voltage source V1 is
the load of this two-port. The voltage V1 (independent quantity) and current I1 are
the parameters of operating or running load regime. Therefore, we use the approach
of Sect. 2.1.1.
As it is known [1, 2], the system of equation of this two-port has the view


I0 Y00 V0  Y10 V1
I1 Y10 V0  Y11 V1 :

4:1

where Y parameters are


Y00 y10 y0 ;

Y11 y10 y1 ;

Y10 y10 :

We determine the characteristic value of regime parameters for the short circuit
SC V1SC 0 and open circuit OC I1OC 0. Then,
V1SC 0;
V1OC

I1SC Y10 V0 ;

Y10
V0 ;
Y11

I1OC 0;

I0SC Y00 V0 ;
I0OC

DY
V0 :
Y11

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_4

4:2
4:3

97

98

4 Two-Port Circuits

I0

TP1

I1
V1

y10

V0
y0

y1

Fig. 4.1 Two-port with a load voltage source

For these parameters, system (4.1) becomes


8
SC
< I0 I0SC  IV1 V1
0

4:4

: I I SC  I1SC V :
1
1
V OC 1
1

Equation (4.4) gives the following input and output load straight lines in Fig. 4.2.
Our circuit represents an active two-pole relatively to load. By denition, the
internal conductivity of the active two-pole is Yi I1SC =V1OC . Therefore, the second
equation of system (4.4) corresponds to Eq. (2.1) of Chap. 2. Then, we may use the
results of this chapter and determine the running regime parameter by an identical
value for the various actual regime parameters (voltages, currents) and for different
sections of circuit (input and output).
As the load characteristics are dened by linear expression (4.4), an afne
transformation or mapping V1 ! I1 ; V1 ! I0 ; I1 ! I0 takes place [3, 9].
Let an initial regime be given by values V11 ; I11 ; I01 . Then, afne ratio (2.6) or (2.8)
has the form
n11

Fig. 4.2 Conformity of the


input and output load straight
lines of a two-port

V1OC  V11
I11  I1OC
I01  I0OC

V1OC  V1SC I1SC  I1OC I0SC  I0OC

4:5

I1, I0
SC

I0
SC

I1

I0

I0

I0(V1)

I1

I1

OC

I0
I1(V1)

V1

V1

OC

V1

V1

4.1 Input-Output Conformity of Two-Ports

99

This afne ratio can be represented by the formal view


n1 I11 I1OC I1SC V11 V1OC V1SC I01 I0OC I0SC :

4:6

Let a subsequent regime be set by values V12 ; I12 ; I02 . Then, the afne ratio is
n21

V1OC  V12
I12  I1OC
I02  I0OC
SC
SC
OC
SC
OC
V1  V1
I1  I1
I0  I0OC

4:7

Now, we use regime change (2.15)


2
1
n21
1 n1  n1

V11  V12
I12  I11
I02  I01

:
V1OC  V1SC I1SC  I1OC I0SC  I0OC

4:8

So, we obtain the identical values for the voltage and currents at the output and
input.
Using (4.8), we get the recalculation formulas
SC
OC
I12 I11 n21
1 I1  I1 ;

SC
OC
I02 I01 n21
1 I0  I0 :

4:9

Let us consider the practical case. Let the voltage V1 be changed once again. We
get regime change (4.8) and currents (4.9). The structure of expression (4.8) shows
that errors of voltage measurement are reduced. Therefore, recalculation formula
(4.9) gives a more great accuracy, than primary Eq. (4.4), which contains the actual
voltage value V1. Also, the invariant value n1 can represent the interest for remote
voltage measurement that will be shown in this chapter further.

4.1.2

Conformity of Cascaded Two-Ports

Let us now consider the cascade connection of two-ports TP1 and TP2 in Fig. 4.3.
A variable voltage source V2 is the load of this connection.
At change of the load voltage V2, we get the family of load straight lines in
Fig. 4.4.
Similarly, we have afne transformations or mapping of regime parameters for
various sections; that is,
V2 ! I2 ;

V 2 ! I1 ;

V2 ! V1 ;

V2 ! I0 ;

I2 ! I1 ! I0 :

Therefore, it is possible to constitute the afne ratio, as an invariant of these


transformations, for initial parameters of all the sections relatively to the load.
Using (4.5), we get

100

4 Two-Port Circuits
TP1

V0, I0

V1, I1

TP 2

I2
V2

y10
y0

y1

Fig. 4.3 Cascade connection of two-ports

Fig. 4.4 Family of load


straight lines for sections of
cascaded two-ports

I2, I1, I0
SC

I0

I0

SC

I0

I1

I0(V1)

SC

I2

OC

I0

I2

I1(V1)

OC

I2

I1
I2(V1)

V2

n12

V2OC  V21
I21  I2OC
SC
OC
SC
V2  V2
I2  I2OC

V1OC  V11
I11  I1OC
SC
OC
SC
V1  V1
I1  I1OC

I01  I0OC
:
I0SC  I0OC

V2

OC

V2

V2

4:10

Also, we may obtain the other afne ratio likewise to (2.7).


Similarly to (4.8), the regime change has the form
n21
2

V21  V22
I22  I21

V2OC  V2SC I2SC  I2OC

V11  V12
I12  I11

V1OC  V1SC I1SC  I1OC

I02  I01
:
I0SC  I0OC

4:11

4.1 Input-Output Conformity of Two-Ports

101

So, we obtain the identical values for the voltages and currents of different
sections. From here, we get the voltage recalculation formula
OC
SC
V12 V11  n21
2 V1  V1

4:12

Example Let us consider a circuit with given values in Fig. 4.5a. The initial and
subsequent regimes correspond to the load voltages V21 1, V22 3. The calculated
currents, voltages of all the sections and their conformity are shown by the diagram
in Fig. 4.5b.
Afne ratio (4.10) for the initial regime
n12

V2OC  V21
3:902  1
0:743;

OC
SC
3:902  0
V2  V2

n12

I21  I2OC
10:43  0
0:743;

I2SC  I2OC 14:03  0

n12

V1OC  V11
4:878  3:86
0:743;

V1OC  V1SC 4:878  3:509

n12

I11  I1OC
15:3  8:78
0:743;

I1SC  I1OC 17:55  8:78

n12

I01  I0OC
34:56  30:49
0:743:

SC
OC
35:97  30:49
I0  I0

Fig. 4.5 a Example of the


cascaded circuit, b conformity
of the regime parameters

(a) V0, I0

(b)

SC

TP1

y10=4

TP 2

V1, I1

y0=1

y0=1

y1=2.4

y1=1

I2

y10=4

V2

OC

V2

V2

V2

V2

3.902

SC

OC

I2

I2

I2

I2

14.03

10.43

3.241

V2

I2

17.55

15.3

10.8

8.78

I1

3.509

3.86

4.562

4.878

V1

35.97

34.56

31.75

30.49

I0

n2

n2

0.7437

0.2317

n2

102

4 Two-Port Circuits

Regime change (4.11)


n21
2

V21  V22
13
0:513:

V2OC  V2SC 3:9  0

Subsequent voltage (4.12)


OC
SC
V12 V11  n21
2 V1  V1 3:86 0:5134:878  3:509 4:562:

4.2
4.2.1

Input-Output Conformity of Two-Ports


as Projective Transformations
Conformity of a Two-Port

Let us consider an asymmetrical two-port TP1 in Fig. 4.6. A variable conductivity


YL1 is the load of this two-port. Therefore, we use the approach of Sect. 2.1.2.
System of Eq. (4.1) has the matrix form


I0
I1

Y00
Y10

  
 Y10
V0

 Y11
V1

4:13

At change of the load conductivity YL1, we get the load straight lines in Fig. 4.7.
In turn, bunches of straight lines with parameters YL1, YIN1 correspond to these
load lines. It is also possible to calibrate the load straight lines in the conductivity
values. The point of maximum load power corresponds to values V1i 0:5V1OC ,
i
Yi . The value Yi is the internal conductivity of circuit relatively to load YL1.
YL1
In previous Sect. 4.1.1 it was shown that the conformity of load straight lines is
an afne transformation for voltage and current. Now, we consider the conformity
YL1 ! YIN1 [5, 9]. For this purpose, we demonstrate the relationship YIN1 YL1 by
the transmission a parameters.

Fig. 4.6 Two-port with a


load conductivity

I0

TP1

V0

I1
V1

y10
y0
YIN1

y1

YL1

4.2 Input-Output Conformity of Two-Ports

103

Fig. 4.7 Input and output


load straight lines

I 1, I 0

SC

I0(V0)

YIN1

SC

YIN1

I0

I1(V1)

YIN1

SC

YL1

I0

SC

I1

YL1

OC

I1

OC

YIN1

I0

YL1

V1

OC

V1

V0

V1, V0

In this case, the equation of two-port has a view




V0
I0

   " Y11
a12
V1

YD10Y
a22
I1
Y

a
11
a21

10

1
Y10
Y00
Y10

#  
V1
:

I1

4:14

The inverse expression




V1
I1

a22

a21

  
a12
V0

a11
I0

4:15

The determinant of a matrix


DA a11 a22  a12 a21 1:
This feature of a parameters allows introducing the hyperbolic functions
DA ch2 c  sh2 c 1;
where is an attenuation coefcient.
Then, Eq. (4.14) is given by


V0

I0
YIN1:C

#
 " YpL1:C

V1
chc shc
DY


1
pI
shc chc
D


4:16

where characteristic admittance at the input and output are


YIN1:C

r
Y00

DY ;
Y11

YL1:C

r
Y11

DY :
Y00

4:17

104

4 Two-Port Circuits

In turn, the relation YIN1 YL1 or admittance transformation has the fractionally
linear view by (4.14)
YIN1

I0
a22 YL1 a21

:
V0 a12 YL1 a11

4:18

Using (4.16), we get the normalized form of this expression


YL1
thc
YIN1
Y
L1:C YL1
:
YIN1:C 1 Y thc
L1:C

4:19

Similarly to (2.23) this conformity YL1 ! YIN1 is a projective transformation.


The projective transformations preserve a cross ratio of four points. Therefore,
1
has the view
analogously to (2.24), the cross ratio m1L1 for initial values YL1
SC
1
i
OC
1
m1L1 YL1
YL1
YL1
YL1
1YL1
Yi 0

4:20

1
YL1
 1 Yi  1
Yi
1 :

1
Yi  0
YL1  0
YL1

The conformity of all the parameters is given by the diagram in Fig. 4.8. Then,
we can constitute the cross ratio for these parameters.
For the input conductivity
SC
1
i
OC
m1L1 YIN1
YIN1
YIN1
YIN1

4:21

1
SC
i
SC
SC
1
YIN1
 YIN1
YIN1
 YIN1
YIN1
 YIN1
OC  i
OC 1
OC ;
1
YIN1  YIN1 YIN1  YIN1 YIN1  YIN1

i
SC OC
is the center of segment YIN1
YIN1 ; that is,
where the point YIN1
i
OC
YIN1
YIN1

Fig. 4.8 Conformity of the


regime parameters

SC
OC
YIN1
 YIN1
:
2

V1 = 0
SC

SC

I1

YL1 =
SC

SC

V1
1

I1

V1

I1

I1
1

YL1
1

YIN1

YIN1

mL1

V1

YL1
i

YL1
2

OC

V1

V1

OC

I1

OC

YL1

OC

I1 =0

YL1 =0

YIN1

YIN1

YIN1

YIN1

mL1

mL1

4.2 Input-Output Conformity of Two-Ports

105

Also, we have
m1L1 0 V11

V1OC OC
V1
V1 OC 1 1 ;
2
V1  V1

4:22

I1SC
I SC  I 1
0 1 1 1 ;
2
I1

4:23

m1L1 I1SC I11

m1L1 I0SC I01 I0i I0OC

I01  I0SC I0i  I0SC


I0SC  I01


;
I01  I0OC I0i  I0OC I01  I0OC

4:24

where the input current I0i is the center of segment I0SC I0OC ; that is,
I0i I0OC

I0SC  I0OC
:
2

1
2
We consider the load change YL1
! YL1
. This change denes the segment
2
1
2
1
YL1 YL1 and segment YIN1 YIN1 on the line I0 V0 . It may be noted that length of
segments is different for usually used Euclidean geometry. However, if the mapping
is viewed as a projective transformation, an invariant or cross ratio is performed,
which denes the same length of segments.
Similar to (2.27)(2.29), this cross ratio has the forms

2
1
m21
L1 1 YL1 YL1 0

1
YL1
;
2
YL1

SC
2
1
OC
m21
L1 YIN1 YIN1 YIN1 YIN1

2
SC
SC
YIN1
 YIN1
Y 1  YIN1
 IN1
m2L1  m1L1 ;
OC
OC
2
1
YIN1  YIN1 YIN1  YIN1

2
1
OC
m21
L1 0 V1 V1 V1

SC 2 1
m21
L1 I1 I1 I1 0

4:25

4:26

V12  0
V11  0

;
V12  V1OC V11  V1OC

4:27

I12  I1SC I11  I1SC


 1
;
I12  0
I1  0

4:28

I02  I0SC I01  I0SC



I02  I0OC I01  I0OC

4:29

SC 2 1 OC
m21
L1 I0 I0 I0 I0

106

4 Two-Port Circuits

Using (4.26), we get the subsequent input conductivity


2
YIN1

SC
OC 21 1
YIN1
YIN1
mL1 mL1
:
1
1 m21
L1 mL1

4:30

Similarly, the subsequent values


V12
I12

V11 V1OC m21


L1
1
V1 m21

1

V1OC
L1

I11 1

I02

4.2.2

4:31

I1SC I11
:
SC 21
 m21
L1 I1 mL1

4:32

1
I0SC I0OC m21
L1 mL1
:
1
1 m21
L1 mL1

4:33

Versions of Conformities, Invariants, and Cross Ratios

There are different versions or variants of cross ratio depending on the choice of the
base points and a unit point. Let us consider expression (4.19). This expression sets
the conformity YL1 ! YIN1 , including negative values, as it is shown in Fig. 4.9 for
the values YL1:C YIN1:C 1.
Version 1
OC
SC
If the base points are YL1
0; YL1
1 and a unit point is YL1 YL1:C 1, the
cross ratio
1
~ 1L1 1 YL1
1 0
m

1
:
1
YL1

Then, for the input conductivity,


~ 1L1
m

1
1
1
YIN1
 thc
1  thc
1
1  Y 1 thc
1
YIN1
1 IN1 :
1 thc 1

thc
YIN1  thc
YIN1  thc 1  thc

-1

-th

1
YL1

-1

th

1
YIN1

Fig. 4.9 Conformity of points of the line YL1 and YIN1

-1/th

1/th

YL1

YIN1

4.2 Input-Output Conformity of Two-Ports

107

1
2
1
2
We express the regime change for YL1
! YL1
and YIN1
! YIN1
as follows

1
Y 2 Y 1 thc
thc IN1 IN1
1
1  Y 2 thc 1  Y 1 thc YL1
2 IN1  1 IN1
2 :
YIN1  thc
YIN1  thc
YL1

~ 2L1  m
~ 1L1
m21
L1 m

It may be noted that the expression for the regime change and the expressions for
the conductivity changes at the input and output are expressed analytically in
different ways, but their numerical values are equal. In this matter, we can choose
the base points and a unit point in such a way as to these expressions are the same
view. This case is presented by the following version 2.
Version 2
We use the characteristic values 1; 1 as the base points. Then, the regime change
2
1
YL1
YL1

2
1
m21 1 YL1
YL1

1
Y 2 1 YL1
1 1 1YL12 YL11
1 L1
:


2 1
1 1
Y 2 Y 1
YL1
YL1
1  L1 2 L11
1YL1 YL1

The obtained expression shows that there is a solid argumentation to introduce


the value of the load change in the form
21
YL1

2
1
YL1
 YL1
:
2
1
1  YL1 YL1

4:34

21
1 YL1
:
21
1  YL1

4:35

Then, the regime change


m21

Similarly, the same regime change and the conductivity change are resulted at
the input
2
1
m21 1 YIN1
YIN1
1

21

YIN1

2
1 YIN1
;
2
1  YIN1

2
1
YIN1
 YIN1
:
2
1
1  YIN1 YIN1

4:36
4:37

Now, we constitute the cross ratio for the initial regime. It is possible to use the
OC
OC
0 as a unit point. This point corresponds to the point YIN1
thc in
point YL1
Fig. 4.9.

108

4 Two-Port Circuits

Fig. 4.10 Movement of the


segment YL1 YIN1 for different
initial values YL1

-1/th

1/th
1

m (YIN1)

-1

-th

th

YL1

YIN1

m (YL1)

Then
1
m1 YL1 1 YL1
0 1

1
1 YL1
;
1
1  YL1

1
m1 YIN1 1 YIN1
thc 1

1
1 YIN1
1 thc
m1 YL1 :

1
1
 thc
1  YIN1

These cross ratio are shown by dash arrows on the superposed axis in Fig. 4.10.
In turn, we can consider the movement of point from the position YL1 to position
YIN1 . This movement determines the segment YL1 YIN1 . The movement of this
segment for different initial values YL1 is shown by arrows in Fig. 4.10.
Then, the points 1 are xed. It is possible to make the cross ratio for the points
YL1 ; YIN1 relatively the xed points. This cross ratio determines the length of
segment YL1 YIN1 and has the view
m YL1 ; YIN1 1 YL1 YIN1 1:
It is known that the property of such a cross ratio; its value does not depend on
running values YL1 ; YIN1 . For simplication, we set YL1 0; YIN1 thc. Therefore,
the cross ratio
m YL1 ; YIN1

1  thc
KPM :
1 thc

It is seen that length of segment YL1 YIN1 equals maximum efciency of


two-port.
Therefore, it turns out that the concept of maximum efciency is already into this
geometric interpretation, and it is not required to think out this denition.
Version 3
Transformation (4.19) is characterized by another invariant too. This expression for
a symmetric two-port takes the form

4.2 Input-Output Conformity of Two-Ports


Fig. 4.11 Rotation of the
radius-vector Y of constant
length at the angle

109

I1,I0

I0(V0)

I1(V1)

SC

YIN1
YIN1

I0

YL1

I1

Y
Y
OC

YIN1
0
0

 
V0
chc
p

shc
I0 DY

V1


 
shc
V1
p
:

chc
I1 DY

V0

V1,V0

4:38

This transformation can be seen as rotation of radius-vector Y of constant length


at angle c in the pseudo-Euclidean plane I, V in Fig. 4.11.
The invariant is given by
V1 2  I1 2 DY V0 2  I0 2 DY :

4:39

This value is the length of radius-vector Y. On the other hand, the trajectory of
rotation is a hyperbola in the Euclidean plane.
Proposed interpretation corresponds to Lorenzs transformation for mechanics of
relative motion [4]. In turn, expression (4.19) corresponds to the rule of addition of
relativistic velocities.

4.2.3

Conformity of Cascaded Two-Ports

Let us now consider the cascade connection of two-ports TP1 and TP2 in Fig. 4.12.
A variable conductivity YL2 is the load of this connection.
At change of the load conductivity YL2, we get the family of load straight lines in
Fig. 4.13. In this case, we have a projective transformation or mapping of load
straight lines and conductivities YL2 ! YL1 ! YIN1 for various sections. This
mapping is shown conditionally by dash-dot lines. According to (4.18), the resultant relationship YIN1 YL2 has the fractionally linear view.
In turn, the resultant matrix of a parameters is equal to product of the matrices
for the rst and second two-ports. Such a group property conrms projective
transformations.

110

4 Two-Port Circuits

Fig. 4.12 Cascade


connection of two-ports with
a load conductivity

V,I
0

TP1

TP 2

V ,I
2 2

V 1 , I1

y10
y1

y0
Y

L2

Y =Y

IN1

Fig. 4.13 Load straight lines


for the cascade connection of
two two-ports

L1

I2, I1, I0

IN2

I1(V1)

I0(V0)
SC

I2(V2)

I0

SC

I1
SC

I0
SC

I2

V1

OC

I2

I0

I1
1

YL2
2

I2

OC

YL2

I1

OC

V1
0

OC

V2

V0

V2, V1, V0

Therefore, by (4.20) and (4.21), we constitute the cross ratio m1L2 for initial
values of regime parameters relatively to the load YL2 .
The conformity of all the parameters is given in Fig. 4.14.
Then, we get
SC
1
OC
m1L2 YL2
YL2
Yi YL2

1
SC
SC
YL2
 YL2
Yi  YL2
Yi

1 ;
OC
OC
1
YL2  YL2
Yi  YL2
YL2

SC
1
i
OC
m1L2 YL1
YL1
YL1
YL1

1
SC
i
SC
YL1
 YL1
YL1
 YL1
OC  i
OC ;
1
YL1  YL1
YL1  YL1

SC
1
i
OC
m1L2 YIN1
YIN1
YIN1
YIN1

1
SC
i
SC
SC
1
YIN1
 YIN1
YIN1
 YIN1
YIN1
 YIN1
OC  i
OC 1
OC :
1
YIN1  YIN1 YIN1  YIN1 YIN1  YIN1

4:40

4.2 Input-Output Conformity of Two-Ports


Fig. 4.14 Conformity of the
regime parameters of
cascaded two-ports

111
1

SC

YL2

SC

V2

YL2 =
V2

SC

I2

SC

V1

SC

I1

SC

YL1

SC

YL2

V2

V2

I2

YL2

I2

I1

V1

I1

YL1

YIN1

YIN1

mL2

I2

OC

V1

OC

I1

V2

I2 =0

V1

YL1

OC

V1
I1

YL1

V2

YL2

OC

I2

V1

OC

YL2 =0

I1

OC

YL1

OC

YL1

YIN1

YIN1

YIN1

YIN1

mL2

mL2

In the last expression,


i
OC
YIN1
YIN1

SC
OC
YIN1
 YIN1
:
2

Using (4.22)(4.24), we may constitute the same cross ratio for all the currents
and voltages. In particular, for the middle section with the parameters V1 ; I1
m1L2 V1SC V11 V1i V1OC

V11  V1SC V1i  V1SC


V11  V1SC


V11  V1OC V1i  V1OC V1OC  V11

I1SC I11 I1i I1OC

I11  I1SC
I11  I1OC

I1i  I1SC
I1i  I1OC

4:41

I1SC  I11
I11  I1OC

where
V1i V1OC

V1SC  V1OC
;
2

I1i I1OC

I1SC  I1OC
:
2

1
2
We consider the load change YL2
! YL2
. This change denes the segments
1
2
1
2
1
YL2 , YL1 YL1 , and YIN1 YIN1 . The cross ratio denes the same length of these
segments

2
YL2

112

4 Two-Port Circuits

SC
2
1
OC
m21
L2 YL2 YL2 YL2 YL2
SC
2
1
OC
YL1
YL1
YL1
YL1

2
SC
1
SC
1
YL2
 YL2
YL2
 YL2
YL2


2  Y OC
1  Y OC
2
YL2
YL2
YL2
L2
L2
2
SC
1
SC
YL1
 YL1
YL1
 YL1
OC  1
OC
2
YL1  YL1
YL1  YL1

SC
2
1
OC
YIN1
YIN1
YIN1
YIN1

4:42

2
SC
SC
YIN1
 YIN1
Y 1  YIN1
 IN1
:
OC
OC
2
1
YIN1  YIN1 YIN1  YIN1

Using (4.27)(4.29), we may constitute the same cross ratio for all the currents
and voltages. In particular, for the middle section with the parameters V1 ; I1 , we get
SC
m21
V12 V11 V1OC
L2 V1

I1SC

I12

I11

I1OC

V12  V1SC V11  V1SC



V12  V1OC V11  V1OC

I 2  I1SC I11  I1SC


21

:
I1  I1OC I11  I1OC

4:43

Using (4.42), we get the subsequent conductivity


2
YL1

SC
1 OC 21
YL1
NL1
YL1 mL2
;
1 m21
1 NL1
L2

1
NL1

SC
1
YL1
 YL1
OC :
1
YL1  YL1

4:44

Similar to (4.33), the subsequent values


V12

1
V1SC V1OC m21
L2 mL2
;
21
1
1 mL2 mL2

I12

1
I1SC I1OC m21
L2 mL2
:
21
1
1 mL2 mL2

4:45

Proceeding from such a geometrical interpretation, it is possible to give the


following added denition [8] of regimes to Chap. 2
Invariance of regimes and their changes for all sections of circuit.
These invariant relationships of a cascaded circuit dene the original conformity principle of regimes.
Therefore, the denition of regime by cross ratios (4.40)(4.43) removes the
indeterminateness in the choice of possible relative expressions for all parameters
and sections of circuit.
Example Let us consider a circuit with given elements in Fig. 4.15.
The rst two-port has the following Y parameters
Y10 y10 4; Y00 y0 y10 5; Y11 y1 y10 6:4; DY Y00 Y11  Y10 2
p
16; DY 4:
Equations (4.14) and (4.18)

4.2 Input-Output Conformity of Two-Ports


TP 1

V0, I0

113

V1, I1

y10=4

TP 2

y10=4

y0=1

y0=1

y1=2.4

y1=1

YIN1

V2, I2

YL2

YL1=YIN 2

Fig. 4.15 Example of the cascaded circuit with the load conductivity

V0
I0

YIN1

   
  
a11 a12
V1
1:6 0:25
V1

;

4 1:25
a21 a22
I1
I1
I0
a22 YL1 a21
1:25  YL1 4
:

V0 a12 YL1 a11 0:25  YL1 1:6

Inverse expression (4.15)




V1
I1

1:25
4

  
V0
0:25

I0
1:6

4:46

For the second two-port





V1
I1

V2
I2

1:25
2:25

1:25
2:25

YIN2 YL1

  
V2
0:25

;
I2
1:25
  
V1
0:25

;
I1
1:25

4:47

I1
1:25  YL2 2:25
:

V1 0:25  YL2 1:25

The resultant transformations of the outputinput




V0
I0

1:6
4

 
1:25
0:25

2:25
1:25
YIN1

   
V2
0:25
2:5625


I2
1:25
7:8125

  
V2
0:7125

;
I2
2:5625

2:5625  YL2 7:8125


:
0:7125  YL2 2:5625

Let the initial and subsequent load conductivities be given as


1
YL2
10:43;

2
YL2
1:08:

4:48

114

4 Two-Port Circuits

Using (4.48), we get


1
YIN1
3:456;

2
YIN1
3:175:

1
V0 34:56;
I01 YIN1

2
I02 YIN1
V0 31:75:

By (4.46), we obtain
V11 3:86;
V12 4:562;

I11 15:3;
I12 10:8;

1
YL1
15:3  3:86 3:963;
2
YL1
10:8  4:562 2:367:

Now, by (4.47)
V21 1;

I21 10:43;

1
YL1
10:43;

V22 3;

I12 3:241;

2
YL1
1:08:

The calculated regime parameters are shown in Fig. 4.16.


Initial regime cross ratio (4.40)
10:43  1 3:597  1 3:597


0:344;
10:43  0
3:597  0
10:43
3:964  5
3:138  5
SC
1
i
OC

0:344;
YL1
YL1
YL1
YL1

3:964  1:8 3:138  1:8


SC
1
i
OC
YIN1
YIN1
YIN1
YIN1

3:456  3:596 3:323  3:596



0:344  1 0:344:

3:456  3:049 3:323  3:049

SC
1
OC
m1L2 YL2
YL2
Yi YL2

m1L2
m1L2

Fig. 4.16 Conformity of the


calculated regime parameters

SC

YL2

SC

OC

YL2

YL2

YL2

YL2

10.43

3.597

1.08

YL2

OC

V2

V2

V2

V2

V2

1.951

3.902

3.509

3.86

4.193

4.562

4.878

V1

17.55

15.3

13.16

10.8

8.78

I1

SC

OC

YL1

YL1

YL1

YL1

YL1

3.964

3.138

2.367

1.8

3.596

3.456

3.323

3.175

3.049

0.344

3.323

V2

YL1

YIN1
mL2

4.2 Input-Output Conformity of Two-Ports

Cross ratio (4.41) for the middle section


m1L2 V1SC V11 V1i V1OC
m1L2 I1SC I11 I1i I1OC

V11  V1SC 3:86  3:509


0:344;

V1OC  V11 4:878  3:86

I1SC  I11 17:55  15:3


0:344:

15:3  8:78
I11  I1OC

Regime change (4.42) for the conductivities


SC 2
1
OC
m21
L2 YL2 YL2 YL2 YL2

1
YL2
10:43
9:66;

2
1:08
YL1

2
SC
1
SC
YL1
 YL1
YL1
 YL1

2  Y OC
1  Y OC
YL1
YL1
L1
L1
2:367  5
3:964  5

4:64  0:478 9:66;

2:367  1:8 3:964  1:8


SC
1
SC
Y 2  YIN1
YIN1
 YIN1
SC
2
1
OC
YIN1
YIN1
YIN1
YIN1
IN1
OC  1
OC
2
YIN1  YIN1 YIN1  YIN1
3:175  3:596 3:456  3:596

3:323  0:344 9:66:

3:175  3:049 3:456  3:049

SC 2
1
OC
m21
L2 YL1 YL1 YL1 YL1

m21
L2

Regime change (4.43) for the middle section


V12  V1SC V11  V1SC

V12  V1OC V11  V1OC
4:562  3:509 3:86  3:509

3:323  0:344 9:66;

4:562  4:878 3:86  4:878


I 2  I1SC I11  I1SC
I1SC I12 I11 I1OC 21

I1  I1OC I11  I1OC
10:8  17:55 15:3  17:55

9:66:

10:8  8:78
15:3  8:78

SC
2 1 OC
m21
L2 V1 V1 V1 V1

m21
L2

The subsequent regime value


1
m2L2 m21
L2 mL2 9:66  0:344 3:323:

We are checking the subsequent conductivity YL1 by (4.44)


2
YL1

SC
1 OC 21
YL1
NL1
YL1 mL2 5 0:478  1:8  9:66
2:367;

1 m21
1 0:478  9:66
1 NL1
L2

1
NL1

1
Y1SC  YL1
0:478:
1  Y OC
YL1
1

115

116

4 Two-Port Circuits

The check of the subsequent values V1 ; I1 by (4.45)


V12
I12

4.3

1
V1SC V1OC m21
3:509 4:878  3:323 19:72
L2 mL2

4:56;

1
1 3:323
4:323
1 m21
m
L2 L2
1
I1SC I1OC m21
17:55 8:78  3:323
L2 mL2
10:8:

21
1
1 3:323
1 mL2 mL2

Use of Invariant Properties for the Transfer


of Measuring Signals

Different sensors of physical values are used for monitoring of technical or natural
objects. For these, usually remote devices, it is necessary to transmit the measuring
signals, for example, over wire lines. Usually, digital signals are used for this
transmission. But such disadvantages as low noise stability, complicated and
expensive equipment, separate wire lines of power supply and communication take
place in digital systems. Therefore, researches and elaborations of easy-to-use
systems for transmitting signals are important.
Invariant properties of two-port networks allow transmitting the measuring
signals, using even the joint or combined wire line for communication and power
supply [5, 12]. As it was shown, the value of afne and cross ratio does not depend
on two-port network (wire line) parameters, accuracy of measuring devices, and
influence of noises. The parameters of this communication wire line, for example,
joint with a power supply line, are dened by both own parameters and current
consumed by other devices.

4.3.1

Transfer of Signals over an Unstable Two-Port

Let us consider the transfer of signals for the following simple examples. More
complex case of transmitting over a joint wire line for communication and power
supply is presented in [10].
Example 1 The value of cross ratio mL1 is accepted as a transmitted analog signal
VS; that is, VS mL1 . The structure of the appliance, designed to this transfer, is
shown in Fig. 4.17a.
In a short time t1 ; t2 ; . . .t5 , when the parameters of this circuit do not change, the
four samples of load conductivity are transmitting by connecting the respective
SC
1
i
OC
; YL1
; YL1
; YL1
. The connection is realized by a multichannel
conductivities YL1
switching and multi-output generator Gen with a pulse period TK. The conductivity
1
1
is the signal or information sample. Therefore, its value YL1
VS is calculated by
YL1
a unit F for the running signal value VS.

4.3 Use of Invariant Properties for the Transfer


Fig. 4.17 a System of an
accurate transmission of
signal by conductivities,
b diagram of the input current

(a)

117
Gen

V0 , I0

OC

YL1 YL1 YL1


I1

F
SC

SC

YL1

VS

-1

VS

OC

I0 , I0, I0, I0

(b)
I0
SC

I0

I0

I0

OC

I0
0

t1 t2 t3 t4 t5

t1+TK

Let us give expression of cross ratio (4.20). In general case


SC 1
i
OC
m1L1 YL1
YL1 YL1
YL1

1
SC
SC
YL1
 YL1
Y i  YL1
 L1
VS :
OC
OC
1
i
YL1  YL1
YL1  YL1

From this we nd, similar to (4.44), the value


1
YL1

SC
i
OC
YL1
VS NL1
YL1
FVS ;
i
1 VS NL1

i
NL1

SC
i
YL1
 YL1
:
OC
i
YL1  YL1

In particular case
1
i
m1L1 1 YL1
YL1
0

i
YL1
VS ;
1
YL1

1
YL1

i
YL1
:
VS

118

4 Two-Port Circuits

Fig. 4.18 System of an


accurate transmission of
signal by voltage sources

Gen

V0, I0
OC

I1

SC

V1

F
SC

I0 , I0, I0

VS

-1

OC

V1

V1

VS

At the input of the circuit, the cross ratio or the signal is calculated by a unit F 1
similar to (4.24). For this purpose, the measured input currents I0SC ; I01 ; I0i ; I0OC ,
shown in Fig. 4.17b, are used; that is,
VS m1L1 I0SC I01 I0i I0OC

I01  I0SC I0i  I0SC



F 1 I01 :
I01  I0OC I0i  I0OC

The structure of this expression shows that measuring errors of currents mutually
are reduced. The process of connection and calculation is repeated every period TK.
If during this time the parameters of two-port network changed, the values of input
currents changed too (for example, the currents decreased in Fig. 4.17b). But the
cross ratio value remains the same.
Example 2 The value of afne ratio n1 is accepted as a transmitted analog signal VS;
that is, VS n1 . The structure of the appliance is shown in Fig. 4.18.
In a short time, the three samples of load voltages are transmitted by connecting the
respective voltage sources V1OC ; V11 ; V1SC . The voltage V11 is the information sample.
Therefore, its value V11 VS is calculated by a unit F for the running signal value VS.
We use expression of afne ratio (2.6) or (4.5)
n11

V1OC  V11
VS :
V1OC  V1SC

From this,
V11 V1OC  VS V1OC  V1SC FVS :

4.3 Use of Invariant Properties for the Transfer

119

A unit F 1 calculates afne ratio (4.5) or the signal


VS n11 I01 I0OC I0SC

I01  I0OC
F 1 I01 :
I0SC  I0OC

The structure of this expression shows that measuring errors of currents mutually
are reduced. The process of connection and calculation is repeated by period TK.
Also, we may use afne ratio (2.7)
VS

V1OC  V11 I01  I0OC


SC
:
V11  V1SC
I0  I01

Then, the information sample


V11

4.3.2

V1OC VS V1SC
:
VS 1

Conductivity Measurement by an Unstable Two-Port

Let us consider the circuit in Fig. 4.19. The action of this appliance is similar to the
system in Fig. 4.17.
1
of a resistive sensor by input
We need to calculate the conductivity value YL1
SC
i
OC
currents. The given conductivities YL1 ; YL1 ; YL1 are reference conductivities.
Using the known cross ratio
SC
1
i
OC
YL1
YL1
YL1

m1L1 YL1

1
SC
i
SC
YL1
 YL1
YL1
 YL1

;
1  Y OC
i  Y OC
YL1
YL1
L1
L1

Fig. 4.19 System of accurate


measurement of conductivity

Gen

V0 , I0

OC

YL1 YL1 YL1


I1

SC

YL1(I0)
OC

I0 , I0, I0, I0

SC

YL1

YL1

120

4 Two-Port Circuits

we nd the value
1
YL1

SC
i
OC 1
YL1
NL1
YL1
mL1
1
YL1
m1L1 ;
i m1
1 NL1
L1

i
NL1

SC
i
YL1
 YL1
:
i  Y OC
YL1
L1

On the other hand, the cross ratio, by the input currents I0SC ; I01 ; I0i ; I0OC , has the view
m1L1 I0SC I01 I0i I0OC

I01  I0SC I0i  I0SC



m1L1 I0 :
I01  I0OC I0i  I0OC

1
Using this expression, we obtain the value YL1
I0 .

4.4

Deviation from the Maximum Efciency of a Two-Port

We use the results of Sects. 1.5.1 and 2.3. Let us consider a more complex case of a
quadratic curve, as the efciency of a two-port in accordance with expression (1.34)
KP KG

1  KG
:
A  KG

4:49

This expression represents a hyperbola in Fig. 4.20 for all area of load changes.
The positive load consumes energy; the maximum power transfer ratio (1.33)
p p
KP A  A  12 :

4:50

Then, the corresponding voltage transfer ratio


KG A 

p
AA  1:

4:51

In turn, the negative load returns energy and we get the corresponding maximum
values
p p
KP A A  12 ;

KG A

p
AA  1:

4:52

Next, we must determine all the characteristic points [6, 11]. Obviously, there
are points B1 ; 0; KP ; 1; A1 ; KP by Fig. 4.20b. These points correspond to points
T 1; 0; KG ; 1; A; KG of the axis KG .

4.4 Deviation from theMaximum Efciency

121

It is possible to take up the points 0; 1 as the base points and the point KG as a
unit point. But visibly, the other characteristic points have to be dened relatively to
these basic points and not depend on the parameter A of comparable two-ports.
Therefore, we must, at rst, dene the possible systems of all the characteristic
points. To do this, we will study a regime symmetry of efciency using the regime
symmetry for the load power of Sect. 2.3.

4.4.1

Regime Symmetry for the Input Terminals

In Fig. 4.20b, the pole S and polar TA1 determine the mapping or symmetry of the
region of power delivery by the voltage source V0 (above of the polar) on the region
of power consumption of this voltage source (below of the polar). The point KP
passes into the point KP . Points B1 ; A1 are the xed base points and KG B1 1,
KG A1 A. A hyperbola point is assigned by the hyperbolic rotation of
radius-vector RS S.

Fig. 4.20 a Efciency of a


two-port for load changes into
the Cartesian coordinates,
b this efciency is as closed
curve in projective
coordinates

KP

(a)

KP

KP

F
1 A

KG

Q K+ K
GF
G

KG

(b)
KG

KP

KP

B1 M

KP
RS

KG

A1

KG

122

4 Two-Port Circuits

Fig. 4.21 Mutual mapping of


the characteristic points and
the correspondence of input
values

Y00

mIN(0)

KG

KG 1

KG

YIN.C

YIN.C YIN

1 1/mIN(0) 0

mIN

DIN

In turn, the pole T and polar SM determine the mapping or symmetry of the
hyperbola relatively to the straight line KP KP or to the points of maximum
efciency.
For the pole T, the following correspondences take place.
We believe harmonic conjugate points T; 0; Q; 1 onto the line TQ or the axis KG.
Therefore, these points correspond to the points KG T 1; 0, KG Q 0:5; 1.
The mutual mapping of the points 0; 1 relatively to Q is shown by arrow in
Fig. 4.21.
Similarly, for the pole S, the following correspondences take place.
We believe harmonic conjugate points S; KP ; M; KP , which correspond to the
points A; KG , 1; KG of the axis KG . The mutual mapping of the points KG ; KG
relatively to A; 1 is shown by arrow in Fig. 4.21.
Let us use the points A; 1 as the base points and KG as a unit point. Therefore,
a running regime point is expressed by cross ratio
A  KG
mIN 1 KG KG A p
AA  1
YIN
1 YIN:C YIN 0
:
YIN:C

4:53

This value is the deviation of KG from the point KG . The correspondence of the
values KG , YIN ; mIN is shown in Fig. 4.21.
Now, it is necessary to check the cross ratios for the points 0, 1
r
A
Y00
mIN 0
;

A  1 YIN:C

r
A1
1
:
mIN 1

A
mIN 0

4:54

The modules of these values are equal to each other if we use the hyperbolic metric

4.4 Deviation from theMaximum Efciency

123

HIN 0 HIN 1 LnmIN 0:


Then, expression (4.53) becomes
HIN LnmIN :
Therefore, we may introduce the normalized distance or relative deviation from
the matched regime
DIN

HIN
LnmIN 

:
HIN 0 LnmIN 0

4:55

The inverse expression


mIN mIN 0DIN
The expansion of this formula by (4.53) and (4.54) gives
A  KG

A  KG

A
A1

DIN =2

YIN

YIN:C

Y00
YIN:C

DIN

4:56

The rst member of these equations is the relative deviation but this deviation is
determined by a value DIN and by parameters of a circuit (the second member).
If it is necessary to set any equal deviation by KG for different circuits, we get
p
KG A  AA  1

4.4.2

A
A1

DIN =2

4:57

Regime Symmetry for the Output or Load

In Fig. 4.22, the pole F and polar TQ, as the axis KG, determine the mapping or
symmetry of the region of power consumption by the load (above of the polar) on
Fig. 4.22 Symmetry of the
efciency relatively to the
axis KG

KP

KP
T

KP

A
KG

Q KG KGF

A1

124

4 Two-Port Circuits

Fig. 4.23 Mutual mapping of


the characteristic points and
the correspondence of load
values

KG KGF 1

YL.C

Y11

KG KG

YL.C YL

1/mL(A) 1 mL(A) 0 mL(A) 1 mL

DL

the region of power return of this load (below of the polar). The point KP passes
into the point KP . The Points 0, 1 are the xed base points.
In turn, the pole T and polar FQ are a complementary system and determine the
mapping or symmetry of the hyperbola relatively to the straight line KP KP or to
the points of maximum efciency.
For the pole T, the following correspondences take place.
We believe harmonic conjugate points T; 0; Q; 1 onto the line TQ. The mutual
mapping of the points T 1; Q relatively to the points 0, 1 is shown by dash
arrow in Fig. 4.23.
Similarly, for the pole F, the following correspondences take place.
We believe harmonic conjugate points F; KP ; Q; KP , which correspond to the
points KGF ; KG , 0:5; KG of the axis KG. The mutual mapping of the points KG ; KG
relatively to 0:5; KGF A=2A  1 is shown by arrow in Fig. 4.23.
Let us use the point KG as a unit point. Therefore, a running regime point is
expressed by cross ratio
r
1  KG
A

mL 0 KG KG 1
A1
KG
4:58
YL
1 YL:C YL 0
:
YL:C
The correspondence of the values KG, YL ; mL is shown in Fig. 4.23.
Now, it is necessary to check the cross ratio for the rest characteristic points, the
points 1; A
r
A
Y11
;

A1
YL:C

mL 1 

mL A 

r
A1
1
:

A
mL 1

4:59

4.4 Deviation from theMaximum Efciency

125

The negative values show that these points disposed into external area of the
base points. We map these points into internal area and obtain the harmonic
conjugate points 0:5; KG F shown by dash arrows.
Similarly to (4.54) and (4.55), we may at once write the following expressions
for deviation DL relatively to the load
1  KG
KG

r 
DL =2
A
A
;

A1
A1

Fig. 4.24 a Deviation for


two hyperboles relatively to
the load, b comparison of
deviations relatively to the
load and input of one two-port

(a)

YL

YL:C

Y11
YL:C

DL

4:60

0.8

DL

0.25
0.5

0.6

1
0

KP 0.4

0.2

A=1.05

0.5 0.25
DL

A=1.25

0.0
0.0

0.2

0.4

0.6

0.8

1.0

KG

(b) 0.8
DL

0.25
0.5

0.6

0.25
1

A=1.05

0.5

KP 0.4

DIN

0.2

0.0
0.0

0.2

0.4

0.6

KG

0.8

1.0

126

4 Two-Port Circuits

If it is necessary to set any equal deviation by KG for different circuits, we get


KG

1
q


A1 A DL =2
A A1

4:61

As an example, the calibration of the hyperboles for two parameters A by deviation DL of (4.61) is given in Fig. 4.24a. The comparison of calibration by DIN of
(4.57) and DL is shown in Fig. 4.24b.

4.5
4.5.1

Effectiveness of Modular Connections


Complementary Knowledge About a Two-Port

At rst, we develop the results of Sect. 4.4. Using Eq. (4.13), we pass on to the
input and output power of a two-port
P0 I0 V0 Y00 V0 2  Y10 V0 V1
P1 I1 V1 Y10 V0 V1  Y11 V1 2 :

4:62

The maximum input power at SC output, when V1 = 0,


P0M Y00 V0 2 :

4:63

Then, the output current


I1SC Y10 V0 :
We will use the Thvenin equivalent generator parameters. The maximum
output voltage at OC output, when I1 = 0,
V1OC

Y10
V0 :
Y11

The maximum output power of the Thvenin equivalent generator


PGM V1OC I1SC

Y10 2
V0 2 :
Y11

4:64

In turn, the known value, as the effectiveness parameter,


A ch2 c

P0M
:
PGM

4:65

4.5 Effectiveness of Modular Connections

127

Taking into account expressions (4.62)(4.65), we obtain the following equation






P1
P0
P0 2
1
:
A 1
P0M
P0M
P0M
It is possible to consider P0M 1. Therefore,
P1 1  P0  A1  P0 2 :

4:66

As it was told, expression (4.49) represents the hyperbola in Fig. 4.20 with the
center S. This center coordinate corresponds to the value
KPS 2A  1:

4:67

Using (4.50) and (4.52), we get the following equalities


KP KPS  2

p
AA  1;

KP KPS 2

p
AA  1:

4:68

Therefore,
KP KP 2KPS :

4.5.2

Parallel Connection of Two Converters

Let us consider two regulable converters or voltage regulators VR1; VR2 with a
common load R1 in Fig. 4.25a. Two-ports TP1; TP2 with the corresponding
effectiveness parameters A1 ; A2 are losses of these regulators are shown in
Fig. 4.25b.
Taking into account the designations of the input and output powers, we form
the following system of the equations
P0 P01 P02
P1 P11 P12 :
We may believe that the corresponding maximum input powers are equal to
P01M P02M 1:
Then, using (4.66), the efciency of this modular connection takes the view
KP

P1 1  P01  A1 1  P01 2 1  P02  A2 1  P02 2

:
P0
P01 P02

4:69

This equation KP P01 ; P02 represents a two-sheeted hyperboloid with the center
S. Therefore, for this surface, we get characteristic regime curves besides characteristic regime points [7]. The center coordinate corresponds to the value

128
Fig. 4.25 a Parallel
connection of two regulable
converters with losses, b its
equivalent circuit

4 Two-Port Circuits

(a)

(b)

KPS

4A1 A2
 1:
A1 A2

4:70

Similarly to (4.67), we may introduce the value


A

2A1 A2
:
A1 A2

4:71

This value determines the total effectiveness parameter of the given modular
connection. In turn, equalities (4.68) take place, where the values KP ; KP are the
total maximum efciency.
We must note that expression (4.71), as the parallel connection of two resistances R1 ; R2 , corresponds to the harmonic conjugacy of four points or three

4.5 Effectiveness of Modular Connections

129

Fig. 4.26 Interpretation of four harmonic conjugate points 0; R1 ; 2R; R2 and 0; A1 ; A; A2

segments. Let us consider this property in details with the aid of Fig. 4.26. The
points A1 ; A2 are harmonic conjugate relatively to the base points 0; A. Also, the
points R1 ; R2 are harmonic conjugate relatively to the base points 0; 2R.
Therefore, we write the following cross ratio
0 A1 A2 A

A1  0 A2  0

1:
A1  A A2  A

Next, we obtain that


A1
A2

:
A1  A A2  A
From here, expression (4.71) follows.
Analogously, we have
0 R1 R2 2R

R1  0
R2  0

1:
R1  2R R2  2R

Further,
R1
R2

:
R1  2R R2  2R
From here, the known formula follows
R

R1 R2
:
R1 R2

130

4 Two-Port Circuits

Fig. 4.27 An intermediate


two-port A12 determines the
interaction of the two
two-ports A1, A2

P01

P11

A1

A12

P0
V0
P02

4.5.3

P12

A2

Connection of Two-Ports with the Interaction

Other variant of modular connection of two two-ports is presented in Fig. 4.27.


Besides the initial two-ports with the effectiveness parameters A1 ; A2 , there is an
intermediate two-port with a parameter A12 . Therefore, the interaction between the
initial two-ports takes place [8].
In this case, a more general expression than (4.69) for the efciency follows
KP

P1 b1  A1 b1 2 b2  A2 b2 2 2Ab1 b2

;
P0
2  b1  b2

where b1 1  P01 ; b2 1  P02 .


This equation represents a two-sheeted hyperboloid with the center S too. The
center coordinate corresponds to the value
KPS 4

A1 A2  A12 2
 1:
A1 A2 2A12

Similar to (4.71), we may introduce the value


A2

A1 A2  A12 2
:
A1 A2 2A12

4:72

Value (4.72) determines the total effectiveness parameter of the given module
connection.
We may also introduce the values
2

 1 A1 A2  A12 ;
A
A2 A12

 2 A1 A2  A12 :
A
A1 A12

4.5 Effectiveness of Modular Connections

131

These values determine the effectiveness parameters for every output terminal.
Therefore, expression (4.71) carries out too
2
1A
2A
A
2 :
A1 A

4.6
4.6.1

Effectiveness Indices of a Two-Port with Variable


Losses
Problems of Energy Indices

Usually, the efciency is used for evaluation of the power supply system effectiveness. But, there are some problems shown in [11].
For example, the efciency value still tells nothing about the effectiveness of an
appliance. Therefore, this efciency value is compared or confronted with the load
power or some reference regime. But, such proposed indices are insufcient to
compare the effectiveness of diverse systems or at change in system parameters.
After all, each power supply is characterized by the power opportunities too; for
example, maximum powers, currents and so on. Therefore, normalized regime
parameters are used. But, for example, the normalized load power does not give
information on the regime effectiveness.

4.6.2

Influence of Losses on the Load Power

We rewrite expression (4.66) for the normalized values; that is,


P1 P0 1  P0  A1  P0 2 :

4:73

Let us consider the dependences P1 P0 for various losses or effectiveness


parameters A in Fig. 4.28a. These dependences represent a bunch of parabolas
passing through the common point PSC
0 1, P1 0 corresponds to SC regime. At
OC regime, we have the powers
P1 0;

2
POC
0 th c

A1
:
A

We note the parabolas with the characteristic effectiveness parameters A. If


A = 1, then the power POC
0 0. If A < 1, our curve gets into the second quadrant
and the two-port, as an active circuit, gives energy. The singular case corresponds
to A 0; we get the tangential straight line and so called the idealized two-port.

132
Fig. 4.28 Load power via the
input power for various
parameters A. a Cartesian
coordinates, b projective
coordinates

4 Two-Port Circuits

P1

(a)

1
A<0

A=0

A3<1
A=1

B1

P0

B2
2

th

A1>1

A2>A1

C2

C1

P1

(b)

1
A4<A3

A=0
A3<1
A=1
A1>1

P0

0
B1

C1

A4
S

One more case corresponds to A 1; the parabola degenerates to the straight line
P0 1. In turn, the load returns power for the region P1 \0.
Change of losses
Let us consider the projective coordinates in Fig. 4.28b. The bunch of our parabolas
represents the bunch of ellipses passing through the other common point P1 1.

4.6 Effectiveness Indices of a Two-Port with Variable Losses

133

The tangential innitely remote straight line corresponds to this point P1 1 and
the point P0 1.
So, we have the two tangential straight lines, which intersect into a point S. The
point S is called a pole; the straight line, corresponding to A 1, is a polar.
These pole and polar determine the symmetry or mapping of points of ellipses of
the region P0 [ 1 onto the region P0 \1. In this case, we have the symmetry of the
points C1, B1 relatively to the xed or base points Q, S. Analogously, the symmetry
of the points C2, B2 takes place.
Thus, these symmetrical points and base points are the harmonic conjugate
points. Similar to (4.72), we get
C1  Q B1  Q C2  Q B2  Q



1:
C1  S B1  S
C2  S B2  S
From here
C2  Q C1  Q B2  Q B1  Q



:
C2  S C1  S
B2  S B1  S
The obtained relationships are the cross ratios of the comparable pairs C1 ; C2 and
B1 ; B2 relatively to the common base points Q, S; that is,
Q C2 C1 S

C2  Q C1  Q

m21
A P1 Q B2 B1 S:
C2  S C1  S

We may consider this value m21


A P1 as a value of the effectiveness parameter
change.
Let us map, for example, the points B1, B2, Q, S onto the axis P0. Then
m21
A P1 Q B2 B1 S P0 Q P0 B2 P0 B1 P0 S:

4:74

If we have executed all the calculation, the nal expression follows [11]
m21
A P1

r
A1 chc1

:
A2 chc2

4:75

Change of load power


Let us consider a given value of the power P0 in Fig. 4.29a. This vertical straight
line intersects all the parabolas with the characteristic effectiveness parameters
A = 0, A = 1, A 1 and a running value A1 . So, we get the corresponding load
powers P1 0, P11 , P1 1, P1 1 in Fig. 4.29b.

134

4 Two-Port Circuits

Fig. 4.29 Load power


changes for a given value of
the input power. a Projective
coordinates, b Cartesian
coordinates

(a)

P1
1
A=0
A=

P0

1
P0

A=1

A1

P1

(b)

A=0
P1(0)

P1(1)
1

P1

A=1

1
0

P0

P0

A1

A=
P1()

Next, we may form the following cross ratio


m1P1 P1 0 P11 P1 1 P1 1:

4:76

According to (4.73),
P11 1  P0  A1 1  P0 2 ; P1 1 P0 1  P0 ; P1 0 1  P0 ; P1 1
1:

4.6 Effectiveness Indices of a Two-Port with Variable Losses

135

Therefore,
m1P1

P11  1  P0
A1 :
P0 1  P0  1  P0

Analogously, for the value A2, we get the load power P21 and
m2P1

P21  1  P0
A2 :
P0 1  P0  1  P0

Therefore, the load power change has the view


m21
P1

m2P1

m1P1

P21
1P0
P11
1P0

1
1

A2
:
A1

4:77

So, there is a strong reason to introduce a specic index in the form


P1
P1
SC
:
1  P0 P0  P0

4:78

The denominator, as the value PSC


0  P0 , shows the load degree of a two-port.
Therefore, this index gives more information about a running regime than simply
normalized load power.

4.6.3

Influence of Losses on the Efciency

Using (4.73), we get the following efciency expression


KP P0

P1 1  P0
1  P0 2

A
:
P0
P0
P0

4:79

Let us consider the dependences KP P0 for various effectiveness parameters


A in Fig. 4.30a.
These dependences represent a bunch of hyperbolas passing through the common point PSC
0 1, P1 0 corresponds to SC regime. At OC regime we have
KP 0;

2
POC
0 th c

A1
:
A

Let us note the hyperbolas with the characteristic effectiveness parameters A. If


A 1, then the power POC
0 0. In this case, the hyperbola degenerates into two

136

4 Two-Port Circuits

(a)

KP

(b)

KP

S 1

A2

A=1

A=0

C2
A=0

C1

B1

A=1

P0

A2
0

B1
1

B2

P0

A1>1 A2>A1

A2

C1

A1>1
Fig. 4.30 Efciency via the input power for various parameters A. a Cartesian coordinates,
b projective coordinates

straight lines, which determine the point S of intersection; the corresponding efciency KP 1. If A = 0, we obtain the hyperbola too.
Change of losses
Let us consider the projective coordinates in Fig. 4.30b. The bunch of our hyperbolas represents the bunch of ellipses passing through the other common point
KP 1. The tangential straight lines correspond to this point KP 1 and the
point PSC
0 1. Therefore, for A 1, the hyperbola degenerates into the innitely
remote straight line 1.
The above point S is the pole too; the straight line, corresponding to A 1, is
the polar. In this case, we have the symmetry of the points C1, B1 relatively to the
innitely remote straight line 1 or the base points Q; S. Analogously, the symmetry
of the points C2 ; B2 takes place.

4.6 Effectiveness Indices of a Two-Port with Variable Losses

(a)

137

KP

S
A=0
A=1

(b)

1
0

P0
=

P0

KP

KP(0)

A=1

A=0

KP(1)
1

KP

P0

P0
A=

A1
A1

KP( )

Fig. 4.31 Efciency changes for a given value of the input power. a Projective coordinates,
b Cartesian coordinates

Similar to (4.74), we may get at once an expression for the effectiveness


parameter change
m21
A KP S B1 B2 Q 0 P0 B1 P0 B2 1

shc1
:
shc2

4:80

The obtained expression differs from (4.75). From here, it follows that the form
of losses change is determined by the initial dependences (4.73) and (4.79).
Change of efciency
Let us consider a given value of the power P0 in Fig. 4.31a. This vertical straight
line intersects all the hyperbolas with the characteristic effectiveness parameters
A = 0, A = 1, A 1 and a running value A1 . So, we get the corresponding
efciency KP 0, Kp1 , KP 1, KP 1 in Fig. 4.31b.

138

4 Two-Port Circuits

Next, we may form the following cross ratio for the value A1,
m1KP KP 0 KP1 KP 1 KP 1:
According to (4.79)
KP 1 1  P0 ;

KP 0

1  P0
;
P0

KP 1 1:

Analogously, for the value A2, we get the value m2KP .


Similar to (4.77), the efciency change has the view
2
1
2
m21
KP mKP  mKP 1  P0 KP

KP2
1P0
KP1
1P0

1
1

KP1

1  A2 sh2 c2

:
1  A1 sh2 c1

So, there is a strong reason to introduce a specic index in the form


KP
:
1  P0
Therefore, this index gives more information about a running regime than the
simply efciency.

References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGraw-Hill,
New York (2009)
2. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
3. Kagan, V.F.: Osnovania geometrii, Chasti II (Geometry Basics. Part II). Gostekhizdat, Moskva
(1956)
4. Kittel, C.K., Knight, W.D., Ruderman, M.A.: Mechanics. Berkeley Physics Course, vol. 1.
McGraw-Hill, New York (1962)
5. Penin, A.: Projectivelyafne properties of resistive two-ports with a variable load.
Tehnicheskaia elektrodinamika 2, 3741 (1991)
6. Penin, A.: Denition of deviation from the matching regime for two-port circuit.
Electrichestvo 4, 3240 (1994)
7. Penin, A.: The relative regime of four-port connected in parallel: geometrical approach.
Electrichestvo 2, 4957 (1997)
8. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
9. Penin, A.: The invariant properties of two-port circuits. World Acad. Sci. Eng. Technol. 3(4),
893899 (2009). http://www.waset.org/publications/5517. Accessed 30 Nov 2014
10. Penin, A., Sidorenko, A.: Method for signal transmission through the direct current line. MD
Patent 536, 12 Dec 2011

References

139

11. Penin, A.: Denition of normalized energy efciency indices of resistive two-port networks.
Probl. Reg. Energ. 2(22), 2037 (2013). http://journal.ie.asm.md/assets/les/m71_2_246.pdf.
Accessed 30 Nov 2015
12. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer,
Dordrecht (2014)

Chapter 5

Paralleling of Limited Capacity Voltage


Sources

5.1

Introduction

The paralleling of lower power voltage sources (converter modules) offers the
well-known advantages over a single, high-power source. The base problem of such
a power supply system is the load-current sharing among the paralleled modules.
Various approaches of current distribution are known [3]. In the simplest droop
method equalizing resistors are used [4, 5, 11], including lossless passive elements
[10]. Usually, the equality of module parameters is provided; that is, open circuit
voltages and internal resistances. Therefore, the distribution of currents means the
equality of these currents.
On the other hand, scatter of module parameters, possible cases of use of primary voltage sources with different capacity determines the non-uniformity distribution of currents.
Therefore, it is natural to understand the uniform loading of sources in the
relative sense when the actual loading corresponds to the capacity of the source.
The analysis of this power supply system by the method of projective geometry has
led to introduction of some concepts for the quantitative representation of operating
regimes [68].

5.2

Initial Relationships

Let us consider voltage sources V1 ; V2 presented in Fig. 5.1. Resistances Ri1 ; Ri2 are
internal resistances of these voltage sources; equalizing resistors Re1 ; Re2 provide
the distribution of currents for a given load resistance R0 .

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_5

141

142

5 Paralleling of Limited Capacity Voltage Sources

Fig. 5.1 Paralleling of


voltage sources

RL1
R i1

R e2

R i2

1.0

12.5

V2

0.2

RL2

0.857

0.5

1.25

V1

I1

R e1

I2

I0

0.833

R0

12.0

6.507

A circuit in Fig. 5.1 is described by the following system of equations:


8
< V1 Ri1 Re1 I1 V0
V Ri2 Re2 I2 V0
: 2
V0 R0 I0 R0 I1 I2 :

5:1

The possible normalized parameters of a load regime (or source loading) of the
rst voltage source look like
m1

RL1
;
Ri1

J1

I1
:
IM1

5:2

Here, the maximum current of the voltage source corresponds to its short circuit
current
IM1

V1
:
Ri1

5:3

Also, such relationships can be rewritten as


RL1  Ri1
;
RL1 Ri1

IM1  I1
;
IM1 I1

IM1
;
I1

and so on.
These expressions are used in different areas of electrical engineering, radio
engineering, and power. Let us note that all these equations represent fractionally
linear expressions and can be interpreted as projective transformations which
possess an invariant. Therefore, all the above-mentioned expressions are equivalent.
Further we will use expressions (5.2).
Since
RL1

V1  Ri1 I1 V1

 Ri1 ;
I1
I1

5.2 Initial Relationships

143

then
m1

RL1
V1
IM1
1

1
 1  1:
J1
Ri1 Ri1 I1
I1

5:4

From here, we get the inverse expression


I1

IM1
:
m1 1

5:5

Similar consideration can be done for the second source


m2

RL2 IM2
1

 1  1;
J2
Ri2
I2
I2

5.3
5.3.1

J2

I2
V2
; IM2
;
IM2
Ri2

IM2
:
m2 1

5:6
5:7

Influence of the Load Value on the Current


Distribution
Analysis of Paralleling Voltage Sources

Let us write an expression, which associates parameters of source loading, in the


form m2 m1 . From (5.1) it follows:
V1  V2 Ri1 Re1 I1  Ri2 Re2 I2 :
So, by (5.5) and (5.7)
I1

IM1
V1

;
m1 1 m1 1Ri1

I2

V2
;
m2 1Ri2

then
V1  V2 Ri1 Re1

V1
V2
 Ri2 Re2
:
m1 1Ri1
m2 1Ri2

From here, we obtain nally


m2

am1 d  a 1
;
m1  d

5:8

144

5 Paralleling of Limited Capacity Voltage Sources

where
V1
V2 Re2
V2
V1 Re1

; d

;
V1  V2 V1  V2 Ri2
V1  V2 V1  V2 Ri1
V1 Re1
V2 Re2
a  d1

:
V1  V2 Ri1 V1  V2 Ri2

5:9

We write now an expression, which associates parameters of source loading, in the


form J2 J1 . So
J1

IM1
1
;

m1 1IM1 m1 1

J2

IM2
1
:

m2 1IM2 m2 1

Using (5.7), we get nally


J2

d1
1
J1 
:
a1
a1

5:10

The plots of dependences (5.8) and (5.10) are presented in Fig. 5.2.
Expression (5.8) corresponds to a hyperbole, and (5.10) corresponds to a straight
line. The desirable operating regime corresponds to straight lines on these plots; that
is, m2 m1 ; J2 J1 . The crossing of this straight line with the hyperbole plot
gives the two points m1 ; m2 of the equal loading of sources. The working area,
when load consumes energy, corresponds to the rst point m1 . The second point
corresponds to the condition when the voltage sources relatively equally consume
energy. Let us determine the points m1 ; m2 . In this case, expression (5.8) leads to
the quadratic equation
m2  d  am  d  a 1 0:
Its solution gives the two roots
m1 d  a 1;

m2 1:

5:11

J 2 1:

5:12

These roots correspond to points of (5.10)


J 1

1
;
d  a2

For the second point J 2 , the currents I2 ; I1 ! 1. Though this case physically is
not feasible, but its mathematical description allows introducing some necessary
characteristics of a circuit.

5.3 Inuence of the Load Value on the Current Distribution


Fig. 5.2 a Correlated
changes of source loading
parameters as a hyperbole.
b Correlated changes of
source loading parameters as
a straight line

145

m2

(a)
m

(1)

(1)

-1
(2)

-1
-a

(b)

(1)

m1

J2
J

(1)

(1)

0
1
d+1

-1
a-1

5.3.2

(1)

J1

Introduction of Two Concepts

The points of equal loading are the xed points of the projective transformation
m1 ! m2 ; J1 ! J2 [1, 2], as it is shown in Fig. 5.3.
Let us consider in detail this geometrical interpretation of transformation (5.8) for
different initial values of the quantities m1 ; m2 at loading change. These quantities
m1 , m2
J1 , J2

-1

J2

J1

m2 m1

(1)

J1
m

(1)

m1

J2 0
m2

Fig. 5.3 Display of the projective transformation of points m1 ! m2 ; J1 ! J2

146

5 Paralleling of Limited Capacity Voltage Sources

dene a line segment, its length (in the usual sense of Euclidean geometry) or
degree of difference of source loading is decreased at its approach to the xed
points. It is obvious that this length for the different circuits will be various.
Thus, it is possible to enter two concepts; the rst of them denes a circuit: how
much the loadings of sources can differ. The second concept denes deviations of
actual loadings from the xed point in the relative form.
In this case, it is possible to compare running regimes of the different circuits.
For introduction of such characteristics we use a number of concepts of projective geometry, applied in Chap. 2. We constitute the following cross-ratio:
m2 m2 m1 m1

m2  m2 m1  m2

;
m2  m1 m1  m1

5:13

where the points m2 ; m1 are the base ones. Also, it is known that the cross-ratio,
concerning the xed points, does not depend on running points m1 ; m2 . Therefore,
we accept m1 1 for simplication of calculations.So, by (5.8)
m2 1

a  1 d  a 1
a;
1d

then
m2 1  m2
m2 1  m1
a  1
a1

KL :

a  d  a 1 d 1

m2 m2 1 1 m1

5:14

Using the values a and d by (5.9), we get


KL

V2 1 Re2 =Ri2
:
V1 1 Re1 =Ri1

5:15

The obtained expression is dened by circuit parameters only. This expression


characterizes the ability of a circuit to equal loading of sources that corresponds to
the rst entered concept. We name this expression as the non-uniformity loading
factor KL .
Obviously, if m2 ! m1 , then KL ! 1 for a given circuit. In general, the factor
KL 6 1. Equation (5.13), taking into account (5.14), allows expressing the
dependence m2 m1 using only the two parameters of a circuit, such as m1 and KL .
Let us represent (5.13) as
KL

m2 1
m1 1

:
m2  m1 m1  m1

5:16

5.3 Inuence of the Load Value on the Current Distribution

147

From here, we get the value


1

m2

KL m
 1 1K
m1 m1
L
1

m
m1  KL1K
L

5:17

1 KL m1 m1 1  KL m1

:
1  KL m1  KL m1

Dependence (5.17) for different values of KL is presented in Fig. 5.4. The bunch of
hyperboles is obtained for KL 6 1. In turn, if KL 1, these hyperboles degenerate
into the straight line m2 m1 .
Let us analyze expression (5.15). We consider V1 V2 .
Then
KL

1 Re2 =Ri2
:
1 Re1 =Ri1

The condition KL 1 leads to the equality


Re2 Re1

:
Ri2 Ri1
Therefore, it is quite possible to put Re2 Re1 0. Thus, if voltage sources have
identical open circuit voltages, they are equally loaded, and it is independent on
their capacity.
Generally, expression (5.14) allows comparing the factors of non-uniformity of
loading of different circuits; to determine the values of equalizing resistors for
necessary value of this factor.

Fig. 5.4 Bunch of


hyperboles for different KL

m2
(1)

(1)

KL>1
KL=1
KL<1

-1
(2)

-1
-a

(1)

m1

148

5.3.3

5 Paralleling of Limited Capacity Voltage Sources

Comparison of a Loading Regime of Different Circuits

At rst, let us obtain the normalized representation of the dependence m2 m1 . For


this purpose, we will consider the cross-ratios for the quantities (or variables) m1
and m2 , using their conformity, according to transformation (5.8). Therefore, the
cross-ratios are equal among themselves
1 m1 m1 d 1 m2 m1 1 ;
where, according to (5.17),
d

KL m1
:
1  KL

The cross-ratio is the relative expression and gives necessary normalizing of


variables. Therefore, any variety of relative expressions for variables m1 and m2 is
excluded.
Let us present each cross-ratio as
1 m1 m1 d

1 m2 m1 1

m1 1
m1

m1  KL1KL
m1 1

m
m1  KL1K
L




m1 1
1

m
m1  KL1K
L

KL
;
1  KL

m2 1
m1 1
m2 1
 1

:
m2  1 m  1 m1 1

Therefore, we have the equation


m2 1
m1 1
KL

:

m1 1 m1  KL m1 1  KL
1KL

The rst member of this expression represents the normalized value and prompts
how to write the similar value in the second member.
Therefore,
KL m 1 1
 1K
m2 1
1 1
L m
:

1
m1 1 mm11 11  1K

5:18

Similarly, we have by (5.10), (5.12) and (5.14)


J2
d 1 J1
d  a2
J1
1  KL
1 


:
1
1
a  1J
a1
KL
J
J KL

5:19

5.3 Inuence of the Load Value on the Current Distribution

149

It should be noted that expressions (5.18) and (5.19) set certainly deviation of
running parameters of loading from the equal loading regime by the normalized
values. But, it is not enough for comparison of deviations for circuits with different
parameters KL . As an example of the most simple relation (5.19), we will show,
why it turns out.
~ L , but with the
Let us consider two circuits with the different parameters KL ; K
1
identical value J 1. The characteristics of circuits are presented in Fig. 5.5.
Loading regimes may be considered identical if conformity of the characteristic
regime points takes place (are shown by arrows in Fig. 5.5) at change of the load. It
follows from the similarity principle [9]. Then, a projective transformation takes
place and it is set by the center at the point 0 and by three pairs of the characteristic
~ B;
~ D.
~ The points D; D
~ coincide among themselves and
regime points A, B, D, and A;
correspond to the xed point J 1 . The point of running regime C should correspond
~ For such a projective transformation, the cross-ratio is carried out
to the point C.
~C
~D
~ B:
~
A C D B A
~ B
~ are the base points, and points D; D
~ are unit ones.
Here, the points A, B and A;
Therefore, this cross-ratio can be accepted as the equal deviation of the running
~ from unit points.
points C; C
Further, we map the points A, C, D, B onto the axis of current J1 . Then, we
obtain the deviation for the rst source

Fig. 5.5 Comparison of the


loading regime of two
different circuits

(1)

J2/J
1

C
0

B
B

1-K
-K

1-K
-K

(1-K)

(1)

J1/J

150

5 Paralleling of Limited Capacity Voltage Sources

D1 0 J1 J 1 1  KL J 1

J1  0
J 1  0
J1


KL :
1
1
1
J1  1  KL J 1
J1  1  KL J
J  1  KL J

5:20

To compare deviations of sources among themselves, it is necessary to express the


deviation for the second source in the same base points. Therefore, we obtain at
once
D2 0 J2 J 1 1  KL J 1

J2
KL :
J2  1  KL J 1

Thus, the deviations include parameters of a circuit and are not simply the normalized values J1 =J 1 ; J2 =J 1 .
We can represent the deviations in the other form
D1

0 J1 J 1
D2

1
d1

J1
a1
a  1J1

:
J1  d 1 1 d 1 d 1J1  1

0 J2 J 1

1
d1

a  1J2
:
d 1J2  1

5:21
5:22

Taking into account conformity (5.4) and (5.6) between various denitions of
parameters of loading regime, the deviations are expressed in the invariant form
through the corresponding cross-ratio for the variables m1 ; m2

D1 0 J1 J 1

D2 0 J2 J 1

1
d 1
1
d1




1 m1 m1 d

1a
:
m2  d

m1  d
1a

;
m1  d
m1  d

5:23

The values of deviations for the characteristic points and running points are presented in Fig. 5.6.
In particular, the deviation D2 for the second xed point, m2 1, is equal to
the parameter KL ; that is,
D2

a1
KL :
d1

5:24

5.3 Inuence of the Load Value on the Current Distribution


Fig. 5.6 Deviations for the
characteristic and running
points

(2)

(1)

151

J1

1
d+1

J2

0
(2)

m1

(2)

(1)

LnK

m,m
1
2

r2

r1

(1)

(2)

Ln1 Ln2

r1

J1, J2

m2

(da+1)

(1)

r2

1, 2

r1, r2

r1, r2

It turns out that such a deviation depends on a circuit parameter. We can exclude
this dependence as follows. We introduce the hyperbolic metrics or distances
r1 Ln D1 ;

r2 Ln D2 :

5:25

The values of these distances are shown in Fig. 5.6.


In particular,
r 1 Ln D1 0;

r 2 Ln D2 Ln KL :

5:26

Only one nonzero and nite value of the distance r 2 is obtained. Therefore, we can
use this value as the scale and introduce the normalized values
r1

r1
Ln D1

;
2
Ln
KL
r

r2

r2
Ln D2

:
2
Ln
KL
r

5:27

Example 1 We use the specic elements in Fig. 5.1. Maximum currents of sources
(5.3)
IM1 10;

IM2 12:

Parameters of loading regimes (5.2) and (5.6)


m1 10:66; J1 0:0857;

m2 13:4; J2 0:0694:

152

5 Paralleling of Limited Capacity Voltage Sources

Source loading regimes (5.8) and (5.10)


m2

29:8m1 5:2
;
m1  34

J2 1:215J1  0:0347:

First xed points (5.11) and (5.12),


m1 5:2;

J 1 0:1613:

Non-uniformity loading factor (5.14), KL 0:8228.


Deviations (5.23)
D1 1:2343;

D2 1:3986:

Hyperbolic distances (5.25)


r1 0:2105;

r2 0:3354:

Scale (5.26), r 2 Ln 0:8228 0:195.


Normalized distances (5.27)
r1

0:2105
1:0795;
0:195

r2

0:3354
1:72:
0:195

All these values are presented in Fig. 5.7.

Fig. 5.7 Example of the


deviations for the
characteristic and running
points

(2)

(1)

J1

0.6113
(2)

(1)

5.2

0.0857

J2
0.0694

m1

m2

10.66

13.4

1/35

34

m,m
1
2

(2)

0.8228

1.2343

1.3986

r1

r2

0.2105

0.3354

(1)

(2)

-0.195

(1)

r1
1

-1.0795

J1, J2

1, 2

r1, r2

r2
-1.72

r1, r2

5.4 Inuence of the Equalizing Resistance on the Current Distribution

5.4
5.4.1

153

Influence of the Equalizing Resistance on the Current


Distribution
Analysis of Paralleling Voltage Sources

Let us write expressions which associate parameters of source loadings in the form
m2 m1 and J2 J1 with the variable equalizing resistor Re1 for a given load. From
(5.1), (5.5) and (5.7) it follows:
V2

V1
V2
R0
Ri2 Re2 R0 :
m1 1Ri1
m2 1Ri2

Then, we obtain
m2 m1

a m1 d a 1
;
m1  d

5:28

where
Re2 R0
V1 R0
; d
 1;
Ri2
V2 Ri1
Re2 R0 V1 R0
ad1

:
Ri2
V2 Ri1
a

5:29

Also, similarly to (5.10), we get


J2 J1 

d1
1
J1
;
a1
a1

5:30

The plots of dependences (5.28) and (5.30) are presented in Fig. 5.8. Expression
(5.28) corresponds to a hyperbole and (5.30) corresponds to a straight line. The
desirable operating regime corresponds to straight lines on these plots; that is,
m2 m1 ; J2 J1 . The crossing of this straight line with the hyperbole gives the
two points m1 ; m2 of the equal loading of sources. The working area when load
consumes energy corresponds to the rst point m1 . The second point corresponds
to the condition when the voltage sources relatively equally consume energy.
Let us nd the xed points m1 ; m2 . In this case, expression (5.28) leads to the
quadratic equation
m2  d am  d a 1 0:

154

5 Paralleling of Limited Capacity Voltage Sources

Fig. 5.8 a Correlated


changes of source loading
parameters as a hyperbole.
b Correlated changes of
source loading parameters as
a straight line

(a)

m2

(1)

(1)

m
1
m2

-1 0
-1

(2)

(b)

(1)

m1

m1

J2

1
1+
J

(1)

(1)

0
J

(1)

1
1+

J1

Its solution gives the two roots


m1 d a 1;

m2 1:

5:31

J 2 1:

5:32

These roots correspond to points of (5.30)


J 1

1
;
da2

For the second J 2 point, the currents I2 ; I1 ! 1.

5.4.2

Introduction of Two Concepts

The points of equal loading are xed points of the projective transformations, for
example, of points m11 ! m12 ; m21 ! m22 , and J11 ! J21 ; J12 ! J22 , shown in Fig. 5.9.

5.4 Inuence of the Equalizing Resistance on the Current Distribution


Fig. 5.9 Display of the
projective transformations of
points m1 ! m2 ; J1 ! J2

-1

155

m1

(1)

J1
1

+1

m2

+1

m2

m1

(1)

J2

m2

J1

J1
1

J2

m1

J2

These transformations are another kind than those for variable load case (5.8) and
(5.10).
Similarly, it is possible to introduce two concepts; one of them denes a circuit:
how much the source loadings can differ. The second concept denes deviations of
actual loadings from the xed point in the relative form.
For introduction of such characteristics, we use also cross-ratio (5.13)
m2 m2 m1 m1

m2  m2 m1  m2

:
m2  m1 m1  m1

We accept m1 1. So, by (5.28), m2 1 a.


Then

m2 m2 1 1 m1

m2 1  m2
a1
K:

1
d1
m2 1  m

5:33

Using (5.29), we get




V2 Ri1
Ri2 Re2
K
1
\0:
V1 Ri2
R0

5:34

The obtained expression is only dened by circuit parameters and characterizes


ability of a circuit to equal loading of sources that corresponds to the rst introduced
. concept. We name this expression as the non-uniformity loading factor K.
The negative value of K shows that points m1 ; m2 are located on different sides
from the xed point m1 , as it is shown in Fig. 5.8a. For example,
m11 [ m1 ; m12 \m1 . For the variable load case, these points are located on the one
side from the xed point and similar factor KL [ 0 by (5.15).

156

5 Paralleling of Limited Capacity Voltage Sources

Similarly to (5.16), we have


K

m2 1
m1 1

:
m2  m1 m1  m1

5:35

From here
1

m2

Km
 1 1K
m1 m1
1

m
m1  K 1K

5:36

1 Km1 m1 1  Km1

;
1  Km1  K m1
1

m
Therefore, d K 1K
.
Dependence (5.36) for different values of K is presented in Fig. 5.10. We obtain
the bunch of hyperboles with the common point (1).
If K K2 1, then

m2

m1 1
1
2 m1 m
1
m1  m 21

dm1 m1
; a d:
m1  d

The points m1 ; m2 are symmetric points concerning m1 and carry the special name
in projective geometry as harmonic conjugate points shown in Fig. 5.11. In particular, the point m1 d corresponds to the point m2 1 and vice versa. Also, the
center S2 will be on the straight line of equal loading. The cross-ratio for harmonic
conjugate points
m2 m1 m2 m1 1:

Fig. 5.10 Bunch of


hyperboles for different K

m2

(1)

(1)

K1 < -1,
1
2
3

S1

>

K2 = -1, 2=
K3 > -1, 3<

S2
S3
m

(1)

m1

5.4 Inuence of the Equalizing Resistance on the Current Distribution

-1

m1

157

(1)

m1

m2

m2

Fig. 5.11 Display of harmonic conjugate points m1 ! m2 ; m2 ! m1

5.4.3

Comparison of a Loading Regime of Different Circuits

Let us obtain the normalized representation of the dependence m2 m1 . For this


purpose, we consider the cross-ratios for the variables m1 and m2 , using their
conformity, according to transformation (5.28). Therefore, the cross-ratios are equal
among themselves
1 m1 m1 d 1 m2 m1 1:
Therefore, similarly to (5.18) and (5.19), we get the same expressions at once
K m1 1
 1K
m2 1
m1 1
:

1
m1 1 mm11 11  1K

5:37

J2
J1
1K
:
1 
K
J 1
J K

5:38

It should be noted, that expressions (5.37) and (5.38) set certainly deviation of running
parameters of loading from the equal loading regime by the normalized values. But it
is not enough for comparison of deviations for circuits with different parameters K. As
an example of the most simple relation (5.38), we will show, why it turns out.
~ but with the
Let us consider two circuits with the different parameters K; K,
1
identical value J 1. The characteristics of circuits are presented in Fig. 5.12;
we are reminding K\0.
As it was noted above, the loading regimes may be considered identical if
conformity of the characteristic regime points takes place (are shown by arrows) at
change of the load. Then, a projective transformation takes place and it is set by the
center at the point 0 and by three pairs of the characteristic regime points A, B, D,
~ B;
~ D.
~ The points D; D
~ coincide among themselves and correspond to the
and A;
1
~ For such
xed point J . The point of running regime C corresponds to the point C.
a projective transformation, the cross-ratio is carried out
~C
~D
~B
~ :
A C D B A

158

5 Paralleling of Limited Capacity Voltage Sources

Fig. 5.12 Comparison of the


loading regime of two
different circuits

J2/J

(1)

1-K
A

-K
1-K
A

-K

D
1

D
C
C

0
1

B
B

(1-K)

J1/J

(1)

Therefore, this cross-ratio can be accepted as the equal deviation of the running
points C, C from unit points. Further, we map the points A; C; D; B onto the axis
of currents J1 . Then, similarly to (5.20), we obtain the deviations for the rst and
second sources at once
D1 0 J1 J 1 1  KJ 1

J1
K;
J1  1  KJ 1

5:39

D2 0 J2 J 1 1  KJ 1

J2
K:
J2  1  KJ 1

5:40

Thus, the deviations include the parameters of a circuit and are not simply the
normalized values J1 =J 1 ; J2 =J 1 .
With the aim to show the presented reasons for introduction of such deviations,
we will consider the special case K 1. Then, expressions (5.38) and (5.39)
became
J2 =J 1 2  J1 =J 1 ;

D1

J1 =J 1
2  J2 =J 1
1

:
1
1
D2
2  J1 =J
J2 =J

5:41

Thus, deviations look like usual proportions and normalized values.


Similarly to (5.21) and (5.22), we can represent the deviations in the other form
D1
D2




0 J1 J 1

1
d1

0 J2 J 1

1
d1




J1
a  1
:
J1  d 1 1 d 1

5:42

J2
a  1
:
J2  d 1 1 d 1

5:43

5.4 Inuence of the Equalizing Resistance on the Current Distribution

159

Similarly to (5.23), the deviations are expressed in the invariant form through the
cross-ratio for the variables m1 ; m2
D1
D2

0 J1 J 1

1
d1

1
d1


0 J2 J




1 m1 m1 d
1a
:

m2  d

1a
;
m1  d

5:44

In particular, for the second xed point m2 1, the deviation D2 K.


We can also use hyperbolic metric (5.25)
r1 Ln D1 ; r2 Ln D2 :

5:45

Then, by (5.26) we get


r 1 Ln D1 0:
Since the value D2 K\0, we cannot use expression r 2 Ln D2 directly.
^ 2 for the point m2 by
Therefore, at rst, we nd the harmonic conjugate point m
the following condition:
1 m
^ 2 m2 d

m2  d
1  d
2
1:
2
^ d m
^ d
m

5:46

Then, we get
^ 2 1 2d:
m
Next, using (5.44), we nd
^ 2 1 a 1 a K:
D
^ 2  d 1 d
m
This value is also obtained from the similar condition to (5.46)
^ 2 D2 1
0 D

^ 2
D
D2

1:

Finally, we get
r 2 Ln K:

5:47

160

5 Paralleling of Limited Capacity Voltage Sources

Using this value as the scale similar to (5.27), we have at once the same normalized
expressions
r1

r1
;
r 2

r2

r2
:
r 2

5:48

The values of deviations for the characteristic points and running points are presented in Fig. 5.13.
If K 1, then D2 1. Using (5.41), we get r1 r2 . In this case, there is
^ 2 1 D1 . From here, the physical sense of
no need to use the scale, because D
value K 1 follows: deviations of sources are equal by quantity, but are
opposite by a sign (for the entire equalizing resistor values that can be useful for
practice).
For this case, the deviations for the characteristic and running points are presented in Fig. 5.14.
Example 2 We consider the circuit in Fig. 5.1 and rewrite the required date of
Example 1.
Maximum currents of sources (5.3)
IM1 10;

IM2 12:

Parameters of loading regime (5.2), (5.6)


m1 10:66;

J1 0:0857; m2 13:4;

(2)

m = 1

m1

(2)

<0

(1)

J2 0:0694:

m2

m1, m2

( 2 ) >0

1, 2

(2)

r1

r2

r1, r2

r1

r2

r1, r2

Fig. 5.13 Deviations of the characteristic and running points for K 6 1

-1

(2)

=1

m1

r1

(1)

1
0

m2

2=1/1

r2=r1

m1, m2

1, 2

r1, r2

Fig. 5.14 Deviations of the characteristic and running points for K 1

5.4 Inuence of the Equalizing Resistance on the Current Distribution

Further, we pass on to our example.


Source loading regimes (5.28) and (5.30)
m2

6:707m1 12:13
;
m1  4:422

J2 0:703J1 0:13:

First xed points (5.31) and (5.32)


m1 12:13;

J 1 0:0761:

Non-uniformity loading factor (5.33), K 1:421.


Deviations (5.44)
D1 1:235;

D2 0:858 1=1:165:

Hyperbolic distances (5.45)


r1 Ln D1 0:211;

r2 Ln D2 0:1526:

The deviation for the second source turns out less than for the rst one.
Scale (5.47)
r 2 Ln 1:421 0:351:
Normalized distances (5.48)
r1

0:211
0:1526
0:601; r2
0:4347:
0:351
0:351

All these values are shown in Fig. 5.15.

m1
4.422

6.707 10.66

(1)

m2

m1, m2

13.4

12.13

1, 2

1.235

0.858

0.351 0.211

-0.1526

r1, r2

r2

r1, r2

-1.421

1.421

r1
0.601

-0.4347

Fig. 5.15 Example of the deviations by equalizing resistance

161

162

5 Paralleling of Limited Capacity Voltage Sources


^

Example 3 The case of a symmetrical circuit, K 1; a d. We consider the


initial circuit in Fig. 5.1 with the xed points m1 12:13; J 1 0:0761. Let us
^

determine the elements Ri1 ; Ri2 of this symmetrical circuit.


Using (5.31) and (5.29), we get
^

m1  1
V1 R0
5:5647
 1;
2
V2 Ri1

Re2 R0
:
Ri2

From here
Ri1

V1 2R0
1:0325;
V2 m1 1

Ri2 2

Re2 R0
1:2052:
m1  1

where R0 ; Re2 ; V1 , and V2 are equal to the given initial values. Then, maximum
currents of sources (5.3)
IM1 12:1065;

IM2 9:9563:

Let the deviation be D1 1:235 and, by (5.41), D2 1=D1 0:8097. Then, by


(5.44),
m1 d

d1
10:8825;
D1

m2 d

d1
13:67:
D2

Normalized currents (5.4) and (5.6)


J1

1
0:0841;
m1 1

J2

1
0:0682:
m2 1

References
1. Frank, J.A.: Schaums outline of theory and problems of projective geometry. McGrawHill,
New York (1967)
2. Glagolev, N.A.: Proektivnaia geometria. (Projective geometry). Nauka, Moskva (1963)
3. Huang, Y., Tse, C.K.: Circuit theoretic classication of parallel connected DCDC
Converters. IEEE Trans. Circuits Syst. I Regul. Pap. 54(5), 10991108 (2007)
4. Irving, B.T., Jovanovi, M.M.: Analysis, design, and performance evaluation of droop
current-sharing method. In: The 15th Annual IEEE Applied Power Electronics Conference
(2000) http://www.deltartp.com/dpel/dpelconferencepapers/06_3.pdf. Accessed 30 Nov 2014
5. Kim, J.W., Choi, H.S., Cho, B.H.: A novel droop method for converter parallel operation.
IEEE Trans. Power Electron. 17(1), 2532 (2002) http://pearlx.snu.ac.kr/Publication/
IEEE0202.pdf. Accessed 30 Nov 2014
6. Penin, A.: Projectivelygeometric approach to parallel connection of electric energy sources.
Tehnicheskaia elektrodinamika 1, 3744 (1992)

References

163

7. Penin, A.: Analysis of paralleling limited capacity voltage sources by projective geometry
method. Sci. World J. (2014). doi:10.1155/2014/359893
8. Penin, A., Sidorenko, A.: Investigation of the effect of an equalizing resistor on the paralleling
voltage sources by projective geometry. Moldavian J. Phys. Sci. 11(12), 124131 (2012)
http://sfm.asm.md/moldphys/2012/vol11/n12/ins_14_penin.pdf Accessed 30 Nov 2014
9. Venikov, V.A.: Theory of similarity and simulation: with applications to problems in electrical
power engineering. Macdonald, London (1969)
10. Villarejo, J.A., De Jodar, E., Soto, F., Jimenez, J.: Multistage high power factor rectier with
passive lossless current sharing. In: The 7th WSEAS International Conference on Circuits,
Systems, Electronics, Control and Signal Processing (CSECS08) (2008)
11. Wang, J.B.: Primary droop currentsharing control of the parallel DC/DC converters system
considering output cable resistance. Adv. Power Electron. (2011). doi:10.1155/2011/713250,
http://www.hindawi.com/journals/ape/2011/713250/

Part II

Multi-port Circuits. Projective


Coordinates of a Point on the Plane
and Space

Chapter 6

Operating Regimes of an Active Multi-port

6.1
6.1.1

Active Two-Port. Afne and Projective Coordinates


on the Plane
Afne Coordinates

Let us give necessary relationships for an active two-port network in Fig. 6.1 with
changeable load voltage sourcesV1 ; V2 .
Taking into account the specied directions of the load currents, this network is
described by the following system of Y parameters equations [1, 5]:
"

I1

I2

Y11

Y12

 " # " SC;SC #


I1
V1
Y12

SC;SC ;
Y22
V2
I2

6:1

where I1SC;SC ; I2SC;SC are the short circuit SC currents of both loads.
The inverse expression is
3 2
3
Y12
I1SC;SC  I1
7
DY 7 6
7
4
5:
Y11 5
SC;SC
I2
 I2
DY

6:2

I2  I2SC;SC 

Y22
DY
I1  I1SC;SC 
V1 ;
Y12
Y12

6:3

I2  I2SC;SC 

Y12
DY
I1  I1SC;SC 
V2 :
Y11
Y11

6:4

Y22
6
6 7 6 DY
4 54
Y12
V2
DY
V1

From here, we obtain

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_6

167

168

6 Operating Regimes of an Active Multi-port

Fig. 6.1 Active two-ports


with changeable load voltage
sources

Equation (6.3) determines a family of parallel load straight lines with a


parameter V1 . Similarly, Eq. (6.4) determines a family of parallel load straight lines
with a parameter V2 . These load straight lines are shown in Fig. 6.2 for characteristic regimes (as the short circuit V1SC 0; V2SC 0 and open circuit V1OC ; V2OC )
and a running regimeV11 ; V21 .
We consider that the load currents dene the rectangular Cartesian system of
coordinates I1 0I I2 . Then, the load voltages correspond to the afne coordinates
V1 0V V2 [2, 6].
The axis V1 equation is dened by expression (6.4) as V2 0:
I2  I2SC;SC 

Y12
I1  I1SC;SC :
Y11

6:5

This straight line passes through the point SC; SC with the coordinates
I1SC;SC ; I2SC;SC or the point 0V . The value Y12 =Y11 corresponds to a slope angle.
Similarly, the axis V2 equation is dened by expression (6.3) as V1 0:

Fig. 6.2 Systems of


Cartesian coordinates I1 0I I2
and afne coordinates
V1 0V V2

6.1 Active Two-Port. Afne and Projective Coordinates

I2  I2SC;SC 

169

Y22
I1  I1SC;SC :
Y12

6:6

The open circuit regime of both loads corresponds to the point OC; OC
with the coordinates I1OC;OC 0; I2OC;OC 0 or the point 0I .
Using (6.2), we get
V1OC

Y22 SC;SC Y12 SC;SC


I

I
;
DY 1
DY 2

6:7

V2OC

Y12 SC;SC Y11 SC;SC


I

I
:
DY 1
DY 2

6:8

We must map the point OC; OC onto the coordinate axes V1 ; V2 by parallel
lines to these axes.
Then, the coordinates or components I1OC;SC ; I2OC;SC correspond to the point V1OC
on the axis V1 . Using (6.7) and (6.1) as V2 0, we get

I1OC;SC

Y11 V1OC

I1SC;SC

I2OC;SC Y12 V1OC I2SC;SC


Y11 Y22 SC;SC Y11 Y12 SC;SC
1

I
;
I1
DY
DY 2
Y12 Y22 SC;SC Y11 Y22 SC;SC
I

I
:
DY 1
DY 2

6:9
6:10

In turn, the coordinates I1SC;OC ; I2SC;OC correspond to the point V2OC on the axis
V2 . Using (6.8) and (6.1) as V1 0, we obtain



2
Y12
Y11 Y12 SC;SC
I
;
I SC;SC
DY 1
DY 2


Y12 Y22 SC;SC
Y11 Y22 SC;SC

I1
 1
:
I2
DY
DY

I1SC;OC Y12 V2OC I1SC;SC


I2SC;OC Y22 V2OC I2SC;SC

6:11
6:12

Let an initial regime be given by values V11 ; V21 or by a point M 1 . Then, the
coordinates I11;SC ; I21;SC dene the point V11 on the axis V1 . Using (6.1) as V2 0,
we have
I11;SC Y11 V11 I1SC;SC ;

6:13

I21;SC Y12 V11 I2SC;SC :

6:14

In turn, the coordinates I1SC;1 ; I2SC;1 dene the point V21 on the axis V2 .
Using (6.1) as V1 0,

170

6 Operating Regimes of an Active Multi-port

I1SC;1 Y12 V21 I1SC;SC ;

6:15

I2SC;1 Y22 V21 I2SC;SC :

6:16

Let us introduce the normalized coordinates for the point M 1 of this running
regime. The point OC; OC is the scale one for the coordinates V1 0V V2 . Then, the
normalized coordinates have the view
n11

V11  0V
V11

;
V1OC  0V V1OC

6:17

n12

V21  0V
V21
OC
:
OC
V2  0V V2

6:18

Also, it is possible to represent the normalized coordinates by the components of


current
n11

n12

I11;SC  I1SC;SC
I1OC;SC

I1SC;SC

I1SC;1  I1SC;SC
I1SC;OC  I1SC;SC

I21;SC  I2SC;SC
I2OC;SC  I2SC;SC
I2SC;1  I2SC;SC
I2SC;OC  I2SC;SC

6:19

6:20

The obtained expressions are similar to expressions (2.6) and (2.9). Also, we
may obtain the other afne ratio similar to (2.7).
The presented approach will be used for analysis of inputoutput transformations
of a circuit with two inputs and two outputs.
Example 1 We consider the following circuit with given parameters in Fig. 6.3.
System of Eq. (6.1) has the view

Fig. 6.3 Example of the


circuit with two load voltage
sources

6.1 Active Two-Port. Afne and Projective Coordinates

"

I1

I2

171

2
3
 " #
I1SC;SC
V1
5


4
Y12 Y22
V2
I2SC;SC

 " # " #
V1
3
1:2 0:2


:
0:2 0:95
V2
2


Y11 Y12

Therefore, Y parameters are the following values:


Y11 y1 y1N 
Y12 y2N

y21N
1:2;
yR

y1N
0:2;
yR

yR y0N yN y1N y2N 5:208;

Y22 y2 y2N 

y22N
0:95:
yR

The SC currents of both loads are


y1N
V0 0:6  5 3;
yR
y2N
Y20 V0 y0N
V0 0:4  5 2:
yR

I1SC;SC Y10 V0 y0N


I2SC;SC

The inverse expression (6.2) is


"

V1
V2


Y22
1


DY
Y12

0:95
1


1:1
0:2

Y12

Y11

2
4

I1SC;SC  I1

I2SC;SC  I2
#
 "
3  I1
0:2

:
1:2
2  I2

3
5

Equation (6.5) of the axis V1 is


Y12
I1  I1SC;SC I2SC;SC
Y11
1
1
 I1  3 2  I1 2:5:
6
6

I2 

Equation (6.6) of the axis V2 is as follows:


I2 

Y22
I1  I1SC;SC I2SC;SC 4:75I1 16:25:
Y12

The obtained coordinates V1 0V V2 are shown in Fig. 6.4.

172

6 Operating Regimes of an Active Multi-port

Fig. 6.4 Example of the


afne coordinates V1 0V V2

Open circuit voltages are (6.7) and (6.8)


Y22 SC;SC
I

DY 1
Y12 SC;SC

DY 1

V1OC
V2OC

Y12 SC;SC 0:95


0:2
3
2 2:9545;
I

1:1
1:1
DY 2
Y11 SC;SC 0:2
1:2
3
2 2:7272:
I2

1:1
1:1
DY

Currents (6.9) and (6.10) are given as


I1OC;SC Y11 V1OC I1SC;SC 1:2  2:9545 3 0:5454;
I2OC;SC Y12 V1OC I2SC;SC 0:2  2:9545 2 2:5909:
Currents (6.11) and (6.12) are given as
I1SC;OC Y12 V2OC I1SC;SC 0:2  2:7272 3 3:5454;
I2SC;OC Y22 V2OC I2SC;SC 0:95  2:7272 2 0:5909:
Let the initial regime be given as follows:
V11 1:958; V21 1:6494:
Currents (6.1) are
I11 Y11 V11 Y12 V21 I1SC;SC 1:2  1:958 0:2  1:6494 3 0:979;
I21 Y12 V11  Y22 V21 I2SC;SC 0:2  1:958  0:95  1:6494 2 0:8247:

6.1 Active Two-Port. Afne and Projective Coordinates

173

Currents (6.13) and (6.14) are


I11;SC Y11 V11 I1SC;SC 1:2  1:958 3 0:6504;
I21;SC Y12 V11 I2SC;SC 0:2  1:958 2 2:3916:
Currents (6.15) and (6.16) are given as
I1SC;1 Y12 V21 I1SC;SC 0:2  1:6494 3 3:3298;
I2SC;1 Y22 V21 I2SC;SC 0:95  1:6494 2 0:433:
Normalized expressions (6.17) and (6.18) are
n11

V11
1:958
OC 2:9545 0:6627;
V1

n12

V21
1:6494
OC 2:7272 0:6047:
V2

We are checking normalized coordinates (6.19) and (6.20) as


n11

I11;SC  I1SC;SC

I21;SC  I2SC;SC

I1OC;SC  I1SC;SC I2OC;SC  I2SC;SC


0:6504  3
2:3916  2

0:6627;

0:5454  3 2:5909  2
I1SC;1  I1SC;SC
I2SC;1  I2SC;SC
n12 SC;OC

I1
 I1SC;SC I2SC;OC  I2SC;SC
3:3298  3
0:433  2

0:6047:

3:5454  3 0:5909  2

6.1.2

Particular Case of a Two-Port. Introduction


of the Projective Plane

Now, let us introduce the projective plane. For this purpose, we consider the most
simple active circuit, which contains two load conductivities YL1 ; YL2 in Fig. 6.5a.
If an internal resistance of voltage source V0 is equal to zero, the independent
change of load currents takes place at the independent change of the load conductivities. The point M of running regime can be set by the values of load conductivities MYL1 ; YL2 or by load currents MIL1 ; IL2 .
The family of load straight lines (there are parallel lines) coincides, for example,
with the rectangular Cartesian coordinates in the Euclidean plane in Fig. 6.5a. Then,
the calibrations of the coordinate axes, by the values of currents and conductivities,
coincide. It is obvious that the circuit does not possess the own scales. Therefore, it
is possible to express the regime only by the absolute or actual values of currents or
conductivities.

174
Fig. 6.5 a Simple active
circuit and its load straight
lines in the Cartesian
coordinates. b This load
straight lines in the projective
coordinates

6 Operating Regimes of an Active Multi-port

(a)

(b)

We can represent the obtained family of load straight lines in the projective plane
[7, 10]. In this case, the innitely large currents and conductivities (innitely remote
points) are located in a nite domain in Fig. 6.5b and form the line of innity 1.
Therefore, the load straight lines represent two bunches of straight lines with the
centers in these innitely remote points.
Now, the internal resistance Ri of the voltage source V0 accepts a nite value in
Fig. 6.6. In this case, the dependent change of load currents takes place. There are
two bunches of load straight lines; the bunch centers are dened by the SC current
IM . Then, the straight line of the maximum current IM passes across these centers.
This straight line is similar to the line of innity 1.
The obtained deformed coordinate grid denes the projective plane. The axes
of coordinates can be calibrated by the values of corresponding currents or conductivities of the loads. In this case, there is an internal scale, that is, the value of
conductivity Yi 1=Ri or the current IM .
It is possible to accept that the load conductivities are equal, for example, to the
internal conductivity Yi and dene the point Mi of characteristic regime.
The projective coordinates are uniquely set by four points; there are three points
of the reference triangle G1 0G2 and scale point Mi [3, 4].
The point M of running regime can be set by the values of load conductivities
(non-uniform coordinates) MYL1 ; YL2 or by the load currents (homogeneous
coordinates) MIL1 ; IL2 ; VL . The sense of homogeneous coordinates consists that
they are proportional to the length of perpendiculars from a point to the sides of
reference triangle. The homogeneous coordinates are used for uncertainty elimination, when the point is on the line of innity 1. The presence of the fourth

6.1 Active Two-Port. Afne and Projective Coordinates

175

Fig. 6.6 Active circuit with


the internal resistance of
voltage source and its load
straight lines in the projective
coordinates

(characteristic) point allows to introduce the cross-ratios mL1 ; mL2 and to set the
regime in the relative form.

6.1.3

General Case of a Two-Port. Projective Coordinates

Let us give necessary relationships for an active two-port network in Fig. 6.7 with
changeable load conductivities YL1 ; YL2 [9, 11]. Taking into account the specied
directions of the load currents, these active two-ports are described by (6.1)
"

I1

I2

 " # " SC #
I1
V1
Y11 Y12


SC ;
Y12 Y22
V2
I2


6:21

where I1SC ; I2SC are SC currents of both loads.


Fig. 6.7 Active two-port
network with changeable load
conductivities YL1 ; YL2

V1

V2

176

6 Operating Regimes of an Active Multi-port

Fig. 6.8 Example of the


circuit with two load
conductivities

Next, we use the circuit of Example 1. In our case, this circuit is shown in
Fig. 6.8. For convenience, we rewrite Y parameters
Y11 y1 y1N 
Y12 y2N

y1N
;
yR

y21N
;
yR

yR y0N yN y1N y2N ;

Y22 y2 y2N 

I1SC Y10 V0 y0N

y1N
V0 ;
yR

y22N
:
yR

I2SC Y20 V0 y0N

6:22
y2N
V0 :
yR

Taking into account the voltages V1 I1 =YL1 ; V2 I2 =YL2 , the equations of two
bunches of straight lines with parameters YL1 ; YL2 are obtained from system (6.21):


DY
V0 Y20 Y12 Y10 Y22  Y12 I2 I1 Y22
;
YL1


DY
:
V0 Y10 Y12 Y20 Y11  Y12 I1 I2 Y11
YL2

6:23

Bunches of these straight lines are presented in Fig. 6.9. The bunch center, a
point G2 , corresponds to the straight lines with the parameter YL1 . We would remind
that bunch center corresponds to such a regime of the load YL1 which does not
depend on its values. It is carried out for the current and voltage I1 0; V1 0 at
the expense of the second load YL2 parameters.
Then, the center G2 parameters will be as follows:
1 SC
I ;
Y12 1


y2
V0 y0N 1
;
y2N

V2G2 
I2G2

Y22 SC
I I2SC
Y12 1

6:24

6.1 Active Two-Port. Afne and Projective Coordinates

G2
YL2

G2
IL2
y2 y2N :
G2
VL2

177

6:25

The center G1 parameters are expressed similarly:


1 SC
I ;
Y12 2


y1
V0 y0N 1
;
y1N

V1G1 
I1G1

Y11 SC
I I1SC
Y12 2
G1
YL1

G1
IL1
y1 y1N :
G1
VL1

6:26
6:27

We accept that the center coordinates dene the characteristic regimes.


Another kind of characteristic regime is the short-circuit regime of both loads
(YL1 1; YL2 1) presented by the point SC in Fig. 6.9. The open-circuit
regime of both loads is also the characteristic regime and corresponds to the origin
of coordinates, the point 0.
So, the reference triangle G1 0 G2 and a unit point SC dene the projective
coordinates.
In turn, Fig. 6.10 demonstrates the axes V1 ; V2 of the familiar afne coordinates.

Fig. 6.9 Two bunches of


straight lines with parameters
YL1 ; YL2

178

6 Operating Regimes of an Active Multi-port

Fig. 6.10 Axes V1 ; V2 form


the afne coordinates

Let an initial or running regime be corresponded to a point M 1 which is set by


1
1
; YL2
and currents I11 ; I21 . Also, this point is dened by the proconductivities YL1
1
jective non-uniform m1 ; m12 and homogeneous n11 ; n12 ; n13 coordinates.
The point 0 is the origin of projective coordinates and the straight line G1 G2 is
the line of innity 1.
The non-uniform projective coordinate m11 is set by the cross-ratio of four points
similar to (2.24):
1
G1
1 YL1

m11 0 YL1

1
1
YL1
10
YL1


:
G1
1  Y G1
1  Y G1
1  YL1
YL1
YL1
L1
L1

6:28

1
G1
YL1
correspond to the base values. The point
There, the points YL1 0; YL1
YL1 1 is a unit point. Also, the values of m1 are shown in Fig. 6.9. For the point
1
G1
YL1
, the projective coordinate m1 1 denes the sense of the line of
YL1
innity G1 G2 .
The projective coordinate m12 is expressed by the same way
1
G2
m12 0 YL2
1 YL2

1
YL2

1
YL2
:
G2
 YL2

6:29

The homogeneous projective coordinates n1 ; n2 ; n3 set the non-uniform coordinates as follows:


n1 qn1

m1 ;
n3 qn3
where q is a proportionality factor.

n2 qn2

m2 ;
n3 qn3

6:30

6.1 Active Two-Port. Afne and Projective Coordinates

179

In turn, the homogeneous coordinates are dened by the ratio of the distances for
points M 1 ; SC to the sides of reference triangle
qn11

d11
I11
SC
;
SC
I1
d1

qn12

d12
I21
SC
;
SC
I2
d2

qn13

d13
:
dSC
3

6:31

For nding the distances d13 ; dSC


to the straight line G1 G2 , the following
3
equation of this straight line is used:
I1
I2
G2  1 0:
G1
I1
I2

6:32

Then, the distance is given as


dSC
3

s
1
1
l 3  2  2 ;
G1
G2
I1
I2



1 I1SC
I2SC


1
;
l3 I1G1 I2G2

where l3 is a normalizing factor.


Further, we use these expressions in the form
 SC

I1
I2SC

 1 l3 dSC
3 ;
I1G1 I2G2


I11
I21

1
I1G1 I2G2

6:33


l3 d13 :

6:34

Then, by (6.31), the homogeneous coordinates have the view


q n11

I11
;
I1SC

q n12

I21
;
I2SC

q n13

I11
I21
1


:
SC
SC
G1
G2
I 1 l 3 d3
I2 l3 d3
l3 dSC
3

6:35

It allows presenting system of Eq. (6.35) in the matrix form:


2

q n1

7 6
6
7 6
6
7 6
6
6qn 7 6
0
6 27 6
7 6
6
5 4
4
1
G1
q n3
I1 l3 dSC
3
I1SC

3 2
0
1
I2SC
1
G2
I2 l3 dSC
3

I1

7 6 7
7 6 7
7 6 7
6 7
0 7
7  6 I2 7 C  I:
7 6 7
1 5 4 5
1
l3 dSC
3

6:36

180

6 Operating Regimes of an Active Multi-port

We do not use the superscripts 1 and 2 for currents. In this equation, the
values I1 ; I2 ; 1 are the homogeneous Cartesian coordinates.
The inverse transformation is
3 2
q I1
I SC
6 q I 7 6 01
6 27 6
7 4 I SC
6
5
4
1
I1G1
q1
2

0
0

I2SC
I2SC
I2G2

3 2n 3
1
6
7 6 n2 7
1
76 7
7 C  n:
5
5
4
SC

l3 d3

6:37

n3

From here, we pass to the non-uniform Cartesian coordinates or currents as


I1

qI1
I1SC m1
I SC
;
I SC
q1
1
m I2G2 m2  l3 dSC
3
I G1 1
1

6:38

qI2
I2SC m2
I2
I SC
:
I SC
q1
1
m I2G2 m2  l3 dSC
3
I G1 1
1

Thus, for preset values of the conductivities YL1 ; YL2 , we nd the coordinates
m1 ; m2 ; transformation (6.38) allows nding the currents I1 ; I2 .
It is possible to represent system of Eq. (6.38) by the normalized or relative
form:
I1
SC
I1G1 I1G1
m1
I1

I SC

1
I1G1

I1SC
I1G1
I2SC
I2G2

m1
I SC

I SC

m2  I1G1  I2G2 1
I1SC
I1G2

m1  1

m1
I2SC
I2G2

m2  1 1

6:39

I SC

2
m
I2
I2G2 2

:
SC
I SC
I2G2 I1G1
m1  1 2G2 m2  1 1

I1

I2

From this system of equations, it is possible to obtain the equations of two


bunches of straight lines, corresponding to (6.23):



I2
I1
1
1  I2SC =I2G2
1  SC G1
1  G2 G1 1 
;
m1
I2
I1
I1 =I1




I1
I2
1
1  I SC =I G1
1  SC1 G21
:
1  G1 G2 1 
m2
I1
I2
I2 =I2

6:40

6.1 Active Two-Port. Afne and Projective Coordinates

181

Also, system of Eq. (6.39) allows obtaining in the relative form the equations of
load characteristics I1 V1 ; V2 ; I2 V1 ; V2 which correspond to Eq. (6.21). For this
purpose, we express the non-uniform coordinates m1 ; m2 by currents and voltages:
m1

YL1
I1 =V1
I1 =I1G1

;
G1
G1
G1
G1
YL1  YL1
I1 =V1  I1 =V1
I1 =I1  V1 =V1G1

m2

I2 =I2G2
:
I2 =I2G2  V2 =V2G2

Substituting these values in system (6.39), we obtain the required equations


I1 V1 ; V2 ; I2 V1 ; V2 , excluding the currents I2 ; I1 accordingly:


I1SC V1
I1SC V2
I1SC
1  G1


;
I1
V1G1 I1G1 V2G2 I1G1


I2
I2SC V1
I2SC V2
I2SC

1


:
I2G2
I2G2 V1G1
I2G2 V2G2 I2G2
I1

I1G1

6:41

The obtained system of equations represents the purely relative expressions.


Therefore, such values, as




I SC I SC I SC
I SC
1  1G1 ; 1G1 ; 2G2 ; 1  2G2 ;
I1
I1 I2
I2
represent the normalized Y parameters.
Let us pass to the absolute values of regime parameters in system (6.41):

 V1
V2
I1 I1G1  I1SC G1  I1SC G2 I1SC ;
V1
V2

 V2
V1
I2 I2SC G1 I2G2  I2SC G2 I2SC :
V1
V2

6:42

From this, it follows that Y parameters are expressed by the parameters of


characteristic regimes:
Y11

I1G2  I1SC
;
V1G1

Y22

I2G1  I2SC
;
V2G2

Y12 

I1SC
I2SC


:
V2G2
V1G1

6:43

182

6 Operating Regimes of an Active Multi-port

Systems of Eqs. (6.39) and (6.40) are important relationships, as represent the
purely relative expressions. It allows dening, at rst, the coordinates m1 ; m2 (as
relative values) for different load conductivities, and then, the normalized currents.
Also, as (6.41), these expressions allow estimating at once a qualitative state of
such a circuit; how much a running regime is close to the characteristic values.
Then, the currents I1G2 ; I2G1 represent the scales. In this sense, initial systems of
Eqs. (6.21) and (6.22), containing the actual or absolute values of Y parameters,
currents and voltages, are uninformative, because they do not give a direct representation about the qualitative characteristics of a circuit.
In addition, it is possible to notice that it is difcult to obtain directly the relative
expressions of type (6.39)(6.43) from systems of Eqs. (6.21) and (6.22). In this
sense, the geometrical interpretation allows to solve this problem easily.
Example 2 We use the date of Example 1.
Equation (6.21) is given as
I1

I2

1:2
0:2

0:2
0:95




V1

V2

Bunch centers (6.24)(6.27) are as follows:


G1
1:5;
I1G1 15; V1G1 10; YL1
G2
I2G2 16:25; V2G2 15; YL2
1:0833

Let the actual parameters of the running regime be given:


1
0:5;
YL1

1
YL2
0:5;

I11 0:979;

I21 0:8247:

Non-uniform projective coordinates (6.28) and (6.29) are


m11 0:25;

m12 0:3158:

The distances are


l3 dSC
3 0:677;

l3 d13 0:8839;

l3 0:0907:

Homogeneous coordinates (6.31) are


q n11 0:979=3 0:3264;

q n12 0:8247=2 0:4123;

q n13 0:8839=0:677 1:3057:

6.1 Active Two-Port. Afne and Projective Coordinates

183

Let us check up the non-uniform coordinates


m11 0:3264=1:3057 0:25;

m12 0:4123=1:3057 0:3158:

Matrix of transformation (6.36) is


2
6
6
6
C 6
6
4

1
3

0


1
15  0:677

1
2
1

16:25  0:677

0
1
0:677

7
7
7
7:
7
5

Matrix of inverse transformation (6.37) is


2

60
C1 6
4 3
15

2
2
16:25

3
7
7:
5

0:677

Let us check up the values of currents and Y parameters:


3  0:25
0:75
0:979;

2
 0:25 16:25  0:3158 0:677 0:7658
2  0:3159
0:825;
I21
0:7658
15  3
16:25  2
Y11
1:2; Y22
0:95;
10
15
3
2
Y12

0:2:
15 10
I11

6.2
6.2.1

3
15

Projective Coordinates in Space


Particular Case of a Multi-port

The popularized power supply system is shown in Fig. 6.11. Three and more loads
are connected to a common supply voltage source V0 or common node N by own
circuits (two-port TP1; TP2 and so on). The interference of loads is observed
because of the internal circuits by y0N ; yN of the voltage source.
Let us use the above approach for an active two-port network with changeable
conductivities of loads.

184

6 Operating Regimes of an Active Multi-port

Fig. 6.11 Power supply


system with a common
voltage source

Fig. 6.12 Active multi-port


with three loads

We now consider the concrete active multi-port network in Fig. 6.12. Similar to
(6.21), this network is described by the following system of equations [8, 11]:
2

I1

Y11
6 7 4
4 I2 5 Y12
Y13
I3

Y12
Y22
Y23

3 2 V 3 2 I SC 3
1
1
Y13
6 7 6 SC 7
5
Y23  4 V2 5 4 I2 5;
Y33
V3
I SC
3

where
Y11 y1 y1N 

yR y0N y1N y2N y3N ;

y1N
y1N
y2
; Y13 y3N
; Y22 y2 y2N  2N ;
yR
yR
yR
2
y2N
y
y3N
; Y33 y3 y3N  3N :
yR
yR

Y12 y2N
Y23

y21N
;
yR

6:44

6.2 Projective Coordinates in Space

185

The short-circuit SC currents of all the loads are


y1N
V0 ;
yR
y3N
Y30 V0 y0N
V0 :
yR

I1SC Y10 V0 y0N


I3SC

I2SC Y20 V0 y0N

y2N
V0 ;
yR

Taking into account the voltages V1 I1 =YL1 ; V2 I2 =YL2 ; V3 I2 =YL3 , the


equations of three bunches of planes are obtained from system of Eq. (6.44) similar
to Eq. (6.23) and the bunches of straight lines of the active two-port in Fig. 6.9. The
intersection of the planes of one bunch among themselves denes a bunch axis. The
equation of the axis of the bunch YL1 corresponds to the condition I1 0; V1 0.
Therefore, this axis is located into the plane I2 ; I3 , as it is shown in Fig. 6.13.
The point of intersection with the current axis I2 is dened by


y2
I2G2 V0 y0N 1
y2N

6:45

that corresponds to the voltage and conductivity of the second load


V2G2 

Y10
y0N
V0 
V0 ;
Y12
y2N

G2
YL2
y2N y2 :

The point of intersection with the current axis I3 denes the third load:


y3
I3G3 V0 y0N 1
;
y3N

V3G3 

y0N
V0 ;
y3N

G3
YL3
y3N y3 :

6:46

Similarly, the axis of the bunch YL2 corresponds to the condition I2 0 and is
located in the plane I1 ; I3 . The point of intersection with the current axis I1 denes
the rst load

Fig. 6.13 Two bunches of


the planes with parameters
YL1 ; YL3

186

6 Operating Regimes of an Active Multi-port

I1G1



y1
V0 y0N 1
;
y1N

V1G1 

y0N
V0 ;
y1N

G1
YL1
y1N y1 :

6:47

The characteristic regime is the short circuit of the loads YL1 1;


YL2 1; YL3 1. The open circuit of the loads is also characteristic and corresponds to the origin of coordinates, the point 0 in Fig. 6.13.
Also, we accept the plane, which passes through three points I1G1 ; I2G2 ; I3G3 on the
axes of coordinates, as the plane of innity 1. This plane and three coordinate
planes I1 0; I2 0; I3 0 form the coordinate tetrahedron 0 G1 G2 G3 .
Let an initial or running regime be corresponded to a point M 1 which is set by
1
1
1
; YL2
; YL3
and currents I11 ; I21 ; I31 . Also, this point is dened by the
conductivities YL1
projective non-uniform coordinates m11 ; m12 ; m13 and homogeneous n11 ; n12 ; n13 ; n14
coordinates, which are set by the coordinate tetrahedron 0 G1 G2 G3 and a unit point
SC. Then, the homogeneous coordinates are dened as the ratio of the distances for
points M 1 ; SC to the planes of coordinate tetrahedron
qn11

I11
;
I1SC

qn12

I21
;
I2SC

qn13

I31
;
I3SC

qn14

d14
:
dSC
4

6:48

The distances are






1 I11
I21
I31
1 I1SC
I2SC
I3SC
SC
d14


1
;
d


1
;
4
l4 I1G1 I2G2 I3G2
l4 I1G1 I2G2 I3G2
s
1
1
1
l4  2  2  2 ;
G1
G2
G3
I1
I2
I3
where l4 is a normalizing factor.
The non-uniform projective coordinate m11 is set by the familiar cross-ratio:
1
G1
m11 0 YL1
1 YL1

1
1
YL1
10
YL1


:
G1
1  Y G1
1  Y G1
1  YL1
YL1
YL1
L1
L1

6:49

1
G1
YL1
correspond to the base values. The point
There, the points YL1 0; YL1
1
G1
YL1 1 is a unit point. For the point YL1 YL1
, the coordinate m1 1
denes the sense of the line of innity G1 G2 . The cross-ratio for m12 ; m13 is
expressed similarly. Therefore, we get the plane G1 G2 G3 of innity 1.
Also, the homogeneous projective coordinates n1 ; n2 ; n3 set the non-uniform
coordinates as follows:

n1 qn1

m1 ;
n4 qn4

n2 qn2

m2 ;
n4 qn4

where q is a proportionality factor.

n3 qn3

m3 :
n4 qn4

6:50

6.2 Projective Coordinates in Space

187

Homogeneous projective coordinates (6.48) have the matrix form


q n C  I;

6:51

where
2

qn1

I1

6 qn 7
6I 7
6 27
6 27
qn 6
7; I 6 7;
4 qn3 5
4 I3 5
2

qn4

1
1

6
6
6
6
6
6
C 6
6
6
6
6
6
4

I1SC

I2SC

I1G1 l4 dSC
4

I2G2 l4 dSC
4

I3SC
1
G3
I3 l4 dSC
4

0


1
l4 dSC
4

7
7
7
7
7
7
7:
7
7
7
7
7
5

The inverse transformation is


q I C1  n;
2 SC
0
I1
6 0 I SC
6
2
C1 6
6 0
0
4 SC
SC
I1
I1G1

I2
I2G2

0
I3SC

0
0

I3SC
I3G3

l4 dSC
4

3
7
7
7:
7
5

6:52

From here, we nd the currents


I1

qI1
SC
q1 I1G1
m1
I
1

qI2
qI3
I2
; I3
:
q1
q1

I1SC m1
I2SC
I2G2

m2

I3SC
I3G3

m3  l4 dSC
4

;
6:53

Then, the normalized expressions for currents are


I SC

1
m
I1
I1G1 1
;

G1
m1 ; m2 ; m3
I1

I SC

2
m
I2
I2G2 2
;

G2
m1 ; m2 ; m3
I2

I SC

3
m
I3
I3G3 3
;

G3
m1 ; m2 ; m3
I3

6:54

188

6 Operating Regimes of an Active Multi-port

where
1
SC
m1 ; m2 ; m3 I1G1
m1  1
I
1

1
I2SC
I2G2

m2  1

I3SC
I3G3

m3  1 1

Also, system of Eq. (6.54) allows obtaining in the relative form the equations of
load characteristics which correspond to Eq. (6.44). For this purpose, we express
the non-uniform coordinates m1 ; m2 ; m3 by the currents and voltages
m1

YL1
I1 =I1G1

;
G1
YL1  YL1
I1 =I1G1  V1 =V1G1

m2

I2 =I2G2
G2
I2 =I2  V2 =V2G2

m3

I3 =I3G3
G3
I3 =I3  V3 =V3G3

Having substituted these values in system (6.54), we obtain the required


equations:


I SC V1
I SC V2
I SC V3
I SC
1  1G1
 1G1 G2  1G1 G3 1G1 ;
G1
I1
V1
I1 V2
I1 V 3
I1


SC
SC
SC
I2
I V1
I
V2
I V3
I SC
 2G2 G1 1  2G2
 2G2 G3 2G2 ;
G2
G2
I2
I2 V1
I2
V2
I2 V3
I2


SC
SC
SC
I3
I3 V 1
I3 V2
I3
V3
I3SC




1


:
I3G3
I3G3 V1G1 I3G3 V2G2
I3G3 V3G3 I3G3
I1

I1G1

6:55

The obtained equations represent the purely relative expressions. The currents
I1G1 ; I2G2 ; I3G3 are the scales. It is obvious that relative expressions (6.55) are generalized to any number of loads by the formal way. If we use the absolute values of
the regime parameters of Eq. (6.54), then Y parameters are expressed by the
parameters of the characteristic regimes.
Example 3 We consider the circuit with given parameters in Fig. 6.14.
System of Eq. (6.44) is
2

I1

1:236
0:17
6 7 4
0:17 0:966
4 I2 5
0:282
0:188
I3

3 2 V 3 2 2:641 3
1
0:282
7
6 7 6
5
0:188  4 V2 5 4 1:761 5:
1:283
2:817
V3

Bunch centers (6.45)(6.47) are


I2G2 16:25;

G2
YL2
1:0833;

G1
1:5:
I1G1 15; YL1

G3
I3G3 14:844; YL3
1:583;

6.2 Projective Coordinates in Space

189

Fig. 6.14 Example of the


circuit with three loads

Let the actual parameters of running regime be given


1
0:5;
YL1

1
YL2
0:5;

I11 0:974;

1
YL3
1;

I21 0:82;

I31 1:61:

Non-uniform projective coordinates (6.49) are


m11 0:25;

m12 0:316;

m13 0:387:

The distances are


l4 dSC
4 0:526;

l4 d14 0:776;

l4 0:0907:

The homogeneous coordinates (6.48) are


q n11 0:974=2:641 0:369;
q n13 0:571;

q n12 0:466;

q n14 1:476:

Matrix of transformation (6.52) is


2

C1

2:641
6 0
6
4
0
0:176

0
0
1:761
0
0
2:817
0:108 0:19

3
0
0 7
7:
0 5
0:526

190

6 Operating Regimes of an Active Multi-port

Let us check current (6.53)


I11

6.2.2

2:641  0:25
0:66

0:974:
0:176  0:25 0:108  0:316 0:19  0:387 0:526 0:67

General Case of a Multi-port. The Balanced Networks

Let us consider the general case of an active multi-port, for example, with three load
conductivities YL1 ; YL2 , and YL3 in Fig. 6.15 [1315]. The circuit is also described
by system of Eq. (6.44).
Therefore, the equations of three bunches of planes are obtained in the form
I1 ; I2 ; I3 ; YL1 0;

I1 ; I2 ; I3 ; YL2 0;

I1 ; I2 ; I3 ; YL3 0:

The equation of the axis of the bunch YL1 corresponds to the condition
I1 0; V1 0 and to equation I2 ; I3 0. Therefore, this axis is located in the
plane I2 ; I3 in Fig. 6.16a.
The points I2 YL1 ; I3 YL1 are the points of intersection with corresponding
axis. Similarly, we obtain the points I1 YL2 ; I3 YL2 of intersection of the bunch
axis YL2 and the points I1 YL3 ; I2 YL3 of intersection of the bunch axis YL3 .
On the other hand, the projective system of coordinates has to represent the
tetrahedron 0 G1 G2 G3 in Fig. 6.16b. In this case, we accept the plane, which passes
through three base points I1G1 ; I2G2 , and I3G3 on the axes of coordinates, as the plane
of innity 1. Therefore, the next conditions have to be satised
I1 YL2 I1 YL3 I1G1 ;
I3 YL1 I3 YL2 I3G3 :

Fig. 6.15 Example of an


active multi-port

I2 YL1 I2 YL3 I2G2 ;

6.2 Projective Coordinates in Space

191

Fig. 6.16 a Points of


intersection do not coincide
among themselves. b Points
of intersection coincide
among themselves and form
the plane of innity 1

(a)

(b)

We must determine requirements for Y parameters. Let us consider the base


point or base values
I1 I1G1 ;

V1 V1G1 :

Then,
I2 0; V2 0;

I3 0;

V3 0:

Using Eq. (6.44), we get


8 G1
< I1 Y11 V1G1 I1SC
0 Y12 V1G1 I2SC
:
0 Y13 V1G1 I3SC :

6:56

From here, the requirements have the view


V1G1

I2SC I3SC

:
Y12 Y13

6:57

192

6 Operating Regimes of an Active Multi-port

Similarly, we consider
I2 I2G2 ;

V2 V2G2 :

Then,
8
< 0 Y12 V2G2 I1SC
I G2 Y22 V2G2 I2SC
: 2
0 Y23 V2G2 I3SC ;

 V2G2

I1SC I3SC

:
Y12 Y23

6:58

I1SC I2SC

:
Y13 Y23

6:59

And we consider
I3 I3G3 ;
8
< 0 Y13 V3G3 I1SC
0 Y V G3 I SC
: G3 23 3 G32 SC
I3 Y33 V3 I3 ;

V3 V3G3 :
 V3G3

Obtained requirements or base points (6.56)(6.59) are formally generalized for


a larger number of loads.
For clarity, let us view the concrete network in Fig. 6.17. This circuit represents a
multi-terminal or distributed network with three loads and N  4 voltage sources. The
voltage sources are connected to each of the loads. Therefore, internal conductivities
y44 ; . . .; yNN of these voltage sources determine a mutual influence of the loads.
We consider the case
I1 I1G1 ;

Fig. 6.17 Example of a


multi-terminal network

V1 V1G1 :

6.2 Projective Coordinates in Space

193

Then, the equalities U4 0; . . .; UN 0 are the particular cases of the


requirements I2 0; V2 0; I3 0; V3 0. Therefore, Eq. (6.56) reduces to the
equalities
I1G1 I4SC    INSC y44 V4    yNN VN ;

V1G1

I4SC
I SC
 N :
y41
yN1

Similar expressions are obtained for the rest loads.


One more circuit important for practice is presented in Fig. 6.18. There is a
distributed network or active:multi-port, for example, with three pairs of voltage
sources and loads, that is, a six-port.
Now, we obtain conditions (6.56)(6.59) with the aid of Fig. 6.19. Let the rst
load voltage be V1 V1G1 .
For the others, we consider
V2 0; I2 0;

V3 0; I3 0:

This condition gives the following relationship:


UaG1

I5SC I6SC

:
yab yac

6:60

In turn, the currents are


I5SC y5b V5 ; I6SC y6c V6 :

Fig. 6.18 Example of a multi-port or distributed network

6:61

194

6 Operating Regimes of an Active Multi-port

Fig. 6.19 Equivalent circuit


of the multi-port for
calculation the base values
I1G1 ; V1G1

Therefore, we get
y5b
y6c
V5
V6 :
yab
yac
We determine the base values as
V1G1 UaG1

I1G1
;
y1a

I1G1 I4 I5SC I6SC Ia ;

G1
YL1

I1G1
;
V1G1

6:62

where
I4 V4 UaG1 y4a ;

Ia ya UaG1 :

Let the second load voltage be V2 V2G2 . For the others, we consider
V1 0; I1 0;

V3 0; I3 0:

This condition gives the following relationship:


UbG2

I4SC I6SC

;
yab ybc

In turn, the current is


I4SC y4a V4 :

6:63

6.2 Projective Coordinates in Space

195

Therefore, we get
y4a
y6c
V4
V6 :
yab
ybc

6:64

From here, we determine the base values


V2G2 UbG2

I2G2
;
y2b

I2G2 I4SC I5 I6SC Ib ;

G2
YL2

I2G2
;
V2G2

6:65

where
I5 V5 UbG2 y5b ;

Ib yb UbG2 :

Let the third load voltage be V3 V3G3 and


V1 0; I1 0;

V2 0; I2 0:

Therefore,
UcG3

I4SC I5SC

:
yac
ybc

From here,
y4a
y5b
V4
V5 :
yac
ybc

6:66

The base values are


V3G3 UcG3

I3G3
;
y3c

I3G3 I4SC I5SC I6 Ic ;

G3
YL3

I3G3
;
V3G3

6:67

where
I6 V6 UcG3 y6c ;

Ic yc UcG3 :

Usually, the voltage sources of distributed power supply system have equal
voltage values and different output powers. Let the voltage source V5 and load YL2 be
the most powerful elements. Therefore, the values of conductivities y5b ; ybc ; yab ; y4a
can be given independently, for example, based on the power effectiveness of
distributed network or current sharing among paralleled voltage sources.
Then, relative to these conductivities, using (6.64) and (6.66), we get the values
of the rest conductivities

196

6 Operating Regimes of an Active Multi-port

y6c

ybc
y4a ;
yab

yac

ybc
yab
y4a
y6c :
y5b
y5b

6:68

We will identify this distributed network as balanced network for output


terminals.
The obtained conductivity values do not limit especially the functional possibility of such circuits but allow simplifying essential calculation of currents.
Let a running regime be corresponded to a point M 1 , which is set by conduc1
1
1
; YL2
; YL3
. Therefore, we may use Fig. 6.13 and relationships (6.48),
tivities YL1
(6.49), (6.52), and (6.53) for nding the load currents I11 ; I21 ; I31 .
Also, we may represent (6.53) in the other form:
I11 I SC

1
I1G1

I SC

1
I1G1

I21

m11

I2SC
I2G2

m12

I2SC
I2G2

I3SC
I3G2

1

I1SC m11
m11  1

I2SC
I2G2

m12  1

qI21
SC
q1 I1G1
m1  1
1

I1

I31

I1SC m11
SC
I
m13  I1G1

I3SC
I3G3

qI31
SC
q1 I1G1
m1  1
1

I1

I3SC
I3G3

m13  1 1

;
6:69

I2SC m12
I2SC
I2G2

m12  1

I3SC
I3G3

m13  1 1

I3SC
I3G3

m13  1 1

I3SC m13
I2SC
I2G2

m12  1

;
:

Therefore,
2

qI11

I1SC

6 17
6 SC 7
6 qI 7
6I 7
6 27
1 6 2 7
qI  6
7 J   6
7 J1   ISC ;
6 qI 1 7
6 I SC 7
4 35
4 3 5
1

q1

1
m11
0

0
m12

6
6
6
J  6
0
0
6
4 1
1
m1  1 G2 m12  1
I1G1 1
I2

0
0

m13
1
I3G3

m13  1

3
0
07
7
7
0 7:
7
5
1

6:70

It is possible to view this expression as a recalculation formula of currents at


respective change of loads from the SC regime to the initial regime M 1 . The values
m11 ; m12 ; m13 are the parameters of such a projective transformation.

6.2 Projective Coordinates in Space

197

Example 4 We consider the network in Fig. 6.18. Let the voltages be


V4 V5 V6 5, the conductivities be y5b 1=2:5; ybc 1=6; yab
1=4; y4a 1=5; ya yb yc 1=20; and y1a y2b y3c 1.
Using (6.68), we get conductivity y6c 1=7:5 and yac 1=12.
We calculate SC output or load currents
I1SC 0:8629;

I2SC 1:246;

I3SC 0:6601:

Voltage source or input SC currents (6.61) are I5SC 2; I6SC 0:666 and current
is I4SC 1.
In turn, voltage value (6.60) is UaG1 8.
Currents (6.63) are I4 13=5 2:6; Ia 8=20 0:4.
Bunch centers (6.62), (6.65), and (6.67) are
I1G1 5:666;

V1G1 13:666;

I2G2 5:466;

V2G2 9:466;

I3G3 5:866;

V3G3 17:866;

G1
YL1
0:4146;
G2
YL2
0:5774;
G3
YL3
0:3283:

The initial regimes are


1
0:2;
YL1

1
YL2
0:25;

I11 0:4232;

1
YL3
0:1666;

I21 0:5673;

I31 0:3350:

Non-uniform projective coordinates (6.49) are


m11 0:3254;

m12 0:3022;

m13 0:3366

The distances are


l4 d14 0:7644;

l4 dSC
4 0:5073:

Homogeneous coordinates (6.48) are


qn11 0:4232=0:8629 0:4904; qn12 0:4553;
qn13 0:5075; qn14 1:5069:
Let us check non-uniform projective coordinates (6.50) as
0:4904
0:3254;
1:5069
0:5075
m13
0:3367:
1:5069

m11

m12

0:4553
0:3022;
1:5069

198

6 Operating Regimes of an Active Multi-port

Inverse matrix (6.52) is


2

C1

0:8629
6 0
6
4 0
0:1523

0
0
1:246
0
0
0:6601
0:2279 0:1125

3
0
0 7
7:
0 5
0:5073

We check currents (6.53) as


0:8629  0:3254
0:1523  0:3254 0:2279  0:3022 0:1125  0:3367 0:5073
0:2808
0:4232;

0:6636
1:246  0:3022
0:6601  0:3367
I21
0:5673; I31
0:335:
0:6636
0:6636
I11

Transformation (6.70) is as follows:


2

qI11

0:3254
6 17
6 qI2 7 6
0
7 6
6
6 174
0
4 qI3 5
0:1191
q1

0
0:3022
0
0:1277

0
0
0:3367
0:1131

2 SC 3
3
I1
0
6 SC 7
6I 7
07
7  6 2 7:
7
05 6
4 I3SC 5
1
1

Then, we may check currents (6.69) as


0:3254  0:8629
0:1191  0:8629  0:1277  1:246  0:1131  0:6601 1
0:2808
0:4232;

0:6636
0:3022  1:246
0:3367  0:6601
I21
0:5673; I31
0:335:
0:6636
0:6636
I11

6.3

Projective Coordinates of an Active Two-Port


with Stabilization of Load Voltages

The popularized power supply system is shown in Fig. 6.20. The low-dropout linear
regulators can be used as voltage stabilizers.
There are important features of such a circuit, that is, the interference of load
currents on the voltage stabilizers regimes takes place; the SC regime has no
physical sense and cannot be accepted as the characteristic regime.

6.3 Projective Coordinates of an Active Two-Port

199

Fig. 6.20 Active two-port network with stabilization of load voltages

This brings up the problem for the choice of characteristic regimes, justication
of the normalized expressions for parameters and equations of circuit, and comparison of regimes between the loads in a given circuit [12].
Let us introduce the projective coordinates using the results of Sect. 6.1.3.
Using Eqs. (6.21)(6.23), we get the following system of equations:
y0N yN y1N
y0N V0  V1 y0N yN  I2 ;
y1N
y0N yN y2N
y0N V0  V2 y0N yN  I1 :
I2
y2N
I1

6:71

These equations dene two bunches of the straight lines I1 ; I2 ; y1N 0;


I1 ; I2 ; y2N 0, with parameters y1N ; y2N . These bunches are presented in
Fig. 6.21. The bunch center G2 corresponds to the straight lines with parameter y1N .
Physically, the bunch center corresponds to such a characteristic regime of the load
Fig. 6.21 Two bunches of
straight lines with parameters
y1N ; y2N

200

6 Operating Regimes of an Active Multi-port

YL1 which does not depend on y1N . It is carried out for the current I1 0 at the
expense of the second load current I2G2 . In this case, the voltage is VN V1 . Using
the rst equation of system (6.71), we obtain the characteristic value of the second
load current
I2G2 y0N V0  V1 y0N yN :

6:72

The characteristic value of the rst load current, which denes the bunch center
G1 of the straight lines with parameter y2N , is expressed similarly
I1G1 y0N V0  V2 y0N yN :

6:73

In this case, the current is given as I2 0, and voltage as VN V2 .


Using the rst equation of system (6.71) and current (6.73), we nd the characteristic value of the rst regulator
G1
yG1
1N I1

y0N yN
y0N V0  V1 y0N yN  I1G1

y0N yN
I1G1 G2
I2  I1G1

I1G1

:
V2  V1

6:74

Similarly, the characteristic value of the second regulator is


G2
yG2
2N I2

y0N yN
I2G2

:
I1G1  I2G2 V1  V2

6:75

1
1
; YL2
or
Let an initial regime corresponds to a point M 1 which is set by loads YL1
1
1
1
1
currents I1 YL1 V1 ; I2 YL2 V2 . The corresponding values of regulator conductivities are dened by system (6.71). However, using (6.72) and (6.73), we obtain
the more convenient relationships as

y11N I11

y0N yN
;
I2G2  I11 I21

y12N I21

y0N yN
:
I1G1  I11 I21

6:76

Also, this point M 1 is dened by the projective nonuniform m11 ; m12 and
homogeneous n11 ; n12 ; n13 coordinates which are set by the reference triangle G1 0 G2
and a unit point. The point 0 is the origin of coordinates and the straight line G1 G2
is the line of innity 1. As a unit point, we must also choose some characteristic
regime using the condition of stability of the voltages V1 ; V2 . As mentioned above,
the SC load current regime is not such one. However, the SC current regime of the
voltage source, when VN 0, allows nding such a characteristic regime. In this
case, the SC current of V0 is

6.3 Projective Coordinates of an Active Two-Port

201

I0M y0N V0 I1 I2 :
This expression corresponds to an equation of straight line. In Fig. 6.21, this line
intersects the coordinate axes in the points I1 I0M ; I2 I0M . The straight lines
SC
with the parameters ySC
1N ; y2N correspond to these points.
SC
Let us determine the values ySC
1N ; y2N . We assume the current I2 0 for the rst
and I1 0 for the second equation of (6.71). Then, the values are
ySC
1N y0N V0 =V1 ;

ySC
2N y0N V0 =V2 :

6:77

Now, using (6.77) and Eq. (6.71), we can dene the load currents
I1SC V1
I2SC

I1G1
;
V2 V1 I1G1 =I0M

I2G2
V2
:
V1 V2 I2G2 =I0M

6:78

These currents correspond to the point SC as a unit point in Fig. 6.21.


Let us now return to the determination of projective coordinates for the running
regime point. The nonuniform projective coordinate m11 is set by cross-ratio (6.28).
In our case, we have
G1
m11 0 y11N ySC
1N y1N

y11N  0
ySC
1N  0

:
G1
SC
1
y1N  y1N y1N  yG1
1N

6:79

SC
The points y1N 0; y1N yG1
1N correspond to the base points. The point y1N is a
unit one. The values of m1 are shown in Fig. 6.21. The nonuniform projective
coordinate m12 is expressed similarly
G2
m12 0 y12N ySC
2N y2N

y12N

y12N
ySC
 SC 2N G2 :
G2
 y2N y2N  y2N

6:80

These expressions demonstrate the conductivities y1N ; y2N in the relative form.
The homogeneous projective coordinates n1 ; n2 ; n3 set the nonuniform coordinates by (6.30). For convenience, we rewrite (6.30) and all the required formulas
qn1
m1 ;
qn3

qn2
m2 :
qn3

6:81

202

6 Operating Regimes of an Active Multi-port

Homogeneous coordinates (6.36) are


2

q n1

3 2

7 6 I1SC
6
7 6
6
7 6
6
6 q n2 7 6
0
7 6
6
7 6
6
5 4
4
1
q n3
I1G1 l3 dSC
3

I1

7 6 7
7 6 7
7 6 7
6 7
0 7
7  6 I2 7 C  I:
7 6 7
1 5 4 5
1
l3 dSC
3

1
I2SC
1
I2G2 l3 dSC
3

6:82

Inverse transformation (6.37) is


3 2
qI I1
I SC
6 q I 7 6 10
6 I 27 6
7 4 I SC
6
5
4
1
I1G1
q 1
2

0
0

I2SC
I2SC
I2G2

3 2n 3
1
6
7 6 n2 7
76 7
C1  n:
5 4 7
5
SC

l3 d3

6:83

n3

From here, we pass to currents (6.38) as


I1

I2

qI I1
I1SC m1
SC
;
qI 1 I1
I2SC
SC
m

m

l
d
1
2
3 3
I1G1
I2G2
qI I2
I2SC m2
SC
:
qI 1 I1
I2SC
SC
m1 G2 m2  l3 d3
I1G1
I2

6:84

Therefore, it is possible to consider that nonuniform and homogeneous projective


coordinates reasonably represent a running regime of circuit by the relative form.
Next, we must obtain the inverse formulas to (6.84) for the calculation of
parameters m11 ; m12 by the load currents. These inverse formulas are obtained at
once by the following method. From (6.81) we get
n1 n3 m1 ; n2 n3 m2 :
Substituting in (6.82), we have
2

q n3 m 1

7 6 I1SC
6
7 6
6
7 6
6
6 q n3 m 2 7 6
0
7 6
6
7 6
6
5 4
4
1
G1
q n3
I1 l3 dSC
3

3 2
0
1
I2SC
1
G2
I2 l3 dSC
3

I1

7 6 7
7 6 7
7 6 7
6 7
0 7
7  6 I2 7 C  I:
7 6 7
1 5 4 5
1
l3 dSC
3

6:85

6.3 Projective Coordinates of an Active Two-Port

203

From here
I1
l dSC
qn3 m1
I1SC 3 3
m1
;

I1
I2
qn3


1
I1G1 I2G2

I2
l dSC
qn3 m2
I2SC 3 3
m2
:

I1
I2
qn3


1
I1G1 I2G2

6:86

Using (6.79), we get


y11N

m11 yG1
1N
;
yG1
1N
1
m1 SC  1
y1N

y12N

m12 yG2
2N
:
yG2
2N
1
m2 SC  1
y2N

6:87

Example 5 Let our circuit be given as the follows:


V0 5;

y0N 2:5;

yN 0:625;

V1 2;

V2 3:

For the rst load, characteristic currents (6.73) and (6.78) are
I1G1 3:125;

I1SC 1:785;

and conductivities (6.74) and (6.77) are


yG1
1N 3:125;

ySC
1N 6:25:

For the second load,


I2G2 6:25;

I2SC 5:357;

yG2
2N 6:25;

Let the initial currents be equal to I11 1;


Then, regulator conductivities (6.76) are
y11N 0:735;

ySC
2N 4:166:

I21 1:

y12N 2:777:

Nonuniform projective coordinates (6.79) and (6.80) are


m11 0:461;

m12 0:154:

The negative values mean that the point M1 and a unit point SC are on the
different sides from the innitely remote straight line G1 G2 .
Distances (6.34) and (6.33) are

204

6 Operating Regimes of an Active Multi-port

l3 d13
l3 dSC
3


1
1

 1 0:52;

3:125 6:25


1:785 5:357

 1 0:428:

3:125
6:25

Homogeneous coordinates (6.35) are


qn11

1
;
1:785

qn12

1
0:186;
5:357

qn13

0:52
:
0:428

Matrix of transformation (6.82) is


2

1
1:785
0

0
6
6
1
6
C 6
5:357
6
4
1
1
3:125  0:428 6:25  0:428

0
0


1
0:428

7
7
7
7:
7
5

Matrix of inverse transformation (6.83) is


2
C

1

6
6
6
6
4

1:785

5:357

1:785
3:125

5:357
0:428
6:25

3
7
7
7:
7
5

Let us check up currents (6.84)


1:785  0:461
0:82
1;

5:357
0:82

0:461


0:153

0:428
 1:785
3:125
6:25
5:357  0:153
1:
I21
0:82

I11

We check parameters (6.86) as


m11
m12

1
1:785 0:428
1
1
3:125 6:25 
1
5:357 0:428
1
1
3:125 6:25 

1
1

0:2397
0:461;
0:52

0:07989
0:154:
0:52

References

205

References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Bennett, M.K.: Afne and Projective Geometry. Wiley, Hoboken (2011)
3. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw
Hill, New York (1967)
4. Glagolev, N.A.: Proektivnaia Geometria. (Projective geometry). Nauka, Moskva (1963)
5. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
6. Penin, A.: Projective equivalency of operating regimes for two-ports. Electrichestvo 8, 4754
(1993)
7. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
8. Penin, A.: Normalized representation of the equations of active multi-port net works on the
basis of projective geometry. Moldavian J. Phys. Sci. 10(34), 350357 (2011). http://sfm.
asm.md/moldphys/2011/vol10/n3-4/index.html. Accessed 30 Nov 2014
9. Penin, A.: Recalculating the load currents of an active multiport with variable parameters on
the basis of projective geometry. Electrichestvo 10, 6673 (2012)
10. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013). https://
sites.google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 Nov 2014
11. Penin, A.: Recalculation of the loads current of active multiport networks on the basis of projective
geometry. J. Circuits Syst. Comput. 22(05), 1350031 (13 pages) (2013) doi: 10.1142/
S021812661350031X. http://www.worldscientic.com/doi/abs/10.1142/S021812661350031X.
Accessed 30 Nov 2014
12. Penin, A.: Comparison of regimes of active two-port networks with stabilization of load
voltages. Inter. J. Electron. Commun. Electr. Eng. 3(6), 118 (2013). https://sites.google.com/
site/ijeceejournal/
13. Penin, A., Sidorenko, A.: Balanced multi-port electric network and its projective coordinates.
Moldavian J. Phys. Sci. 14(12), 102112 (2015)
14. Penin, A., Sidorenko, A.: Sistemul distribuit de alimentare cu energie. MD Patent application
S20150047, 2015
15. Penin, A., Sidorenko, A., Donu, S.: Metod de transmitere a trei semnale prin linia de
comunicaie cu patru re. MD Patent application S20150082, 2015

Chapter 7

Recalculation of Load Currents


of Active Multi-ports

7.1
7.1.1

Recalculation of Currents for the Case of Load


Changes
Active Two-Port

Let us continue with the matter we began in Sect. 6.1.3. Let a subsequent regime
2
2
; YL2
and currents I12 ; I22 of loads.
correspond to a point M 2 with conductivities YL1
The non-uniform coordinates are dened similarly to (6.28) and (6.29)
m21

2
YL1

2
YL1
;
G1
 YL1

m22

2
YL2

2
YL2
:
G2
 YL2

Therefore, the regime change m21


1 is naturally expressed by the cross-ratio [3]
2
1
G1
m21
1 0 YL1 YL1 YL1

m21
2

2
YL2

1
YL2

G2
YL2

2
YL1
0
Y1  0
 1 L1 G1 m21  m11 ;
G1
2
YL1  YL1 YL1  YL1

m22

7:1

m12 :

Using (6.31) and (6.34), we dene the homogeneous coordinates of the point M 2
q n21

I12
I22
d23
SC
; q n22 SC
; q n23 SC
; l3 d23
I1
I2
d3


I12
I22

1 :
I1G1 I2G1

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_7

207

208

7 Recalculation of Load Currents of Active Multi-ports

Fig. 7.1 Regime change at


the expense of change of the
load conductivities

I2
G2

YL2
21

m2

I2

YL2
1

I2
0

G1
2

I1

I1

YL1

YL1

21

m1

The corresponding regime change is shown by arrows in Fig. 7.1.


Using transformations (6.28), we obtain the subsequent currents
I12

I1SC
I1G1

I1SC m21
;
I SC
m21 2G2 m22  l3 dSC
3
I2

I22

I1SC
I1G1

I2SC m22
:
I SC
m21 2G2 m22  l3 dSC
3
I2

Taking into account (7.1), we present the non-uniform coordinates m21 ; m22 as
m21 m21
1

n11
;
n13

m22 m21
2

n12
:
n13

Therefore, the currents


n1

I12

I SC

1
I1G1

m21
1

n11
n13

1
I1SC m21
1 n1
 SC 3 
I
I2G2 m21
2

n11
n13

2
I2SC m21
2 n1
 SC 3 
I
I2G2 m21
2

n12
n13

 l3 dSC
3

n12
n13

 l3 dSC
3

n1

I22 I SC

1
I1G1

m21
1

Then, by (6.37), we get transformation


qI2  C21 1  n1 ;

I1

7.1 Recalculation of Currents for the Case of Load Changes

209

where the matrix


2

C21 1

I1SC m21
1
6 0
4
I1SC
I1G1

m21
1

0
I2SC m21
2
I2SC
I2G2

0
0
l3 dSC
3

m21
2

3
7
5:

Using (6.36), we obtain the resultant transformation as the multiplication of the


two matrixes
qI2  C21 1  C  I1  J21   I1 ;

7:2

where the matrix


2

m21
1
6
J 21  4 0
1
m21
1  1
I G1
1

0
m21
2
1
m21
2  1
I G2

3
0
07
5:
1

From here, we pass to the required currents


I12
I22

q I12

q1

q I2
2
q1

I11 m21
1
I11
I1G1

m21
1  1

I21
I2G2

m21
2  1 1

I21 m21
2
I11
I1G1

m21
1  1

I21
I2G2

m21
2  1 1

;
7:3
:

The obtained relationships carry out the recalculation of currents at a respective


change of load conductivities. These relations are the projective transformations and
possess group properties. The performance of the group properties is advantage of
projective transformations.
32
Let the regime be changed once again; we have changes m32
1 ; m2 . Then, the
nal regime value is expressed as follows:
2
32
21
1
31
1
3
31
1
m31 m32
1  m1 m1  m1  m1 m1  m1 ; m2 m2  m2 :

Using (7.2), we obtain the resultant transformation


qI3  J32   I2  J32   J21   I1  J31   I1 ;
where the matrices
2

m32
1
6 0
32
J  4 32

m1 1
I1G1

0
m32
2

m32
2 1
I2G2

3
0
07
5;
1

m31
1
6 0
31
J  4 31

m1 1
I1G1

0
m31
2

m31
2 1
I2G2

3
0
07
5:
1

210

7 Recalculation of Load Currents of Active Multi-ports

21
Projective transformation (7.2) or (7.3) with parameters m21
1 ; m2 translates any
initial points of the plane I1 ; I2 into a new position shown in Fig. 7.1. The xed
points and xed straight lines are shown too.

Example 1 We continue Example 2 of Sect. 6.1.3 and rewrite the parameters of an


initial regime
1
1
YL1
0:5; YL2
0:5; I11 0:979; I21 0:8247;

m11 0:25; m12 0:3158:


Let the parameters of a subsequent regime be given as
2
2
YL1
1; YL2
2; I12 1:434; I22 1:5504:

The non-uniform coordinates


m21 0:4; m22 0:6486:
Regime change (7.1)
m21
1 0:4  0:25 1:6;

m21
2 0:6486  0:3158 2:0538:

Transformation (7.2)
2

qI12

1:6
0
6 27 4

0
2:0538
4 qI2 5
0:04 0:0648
q1

3 2 I1 3
1
0
6 7
0 5  4 I21 5:
1
1

Subsequent currents (7.3)


qI12
1:6  0:979
1:5664

1:4335;

0:04

0:979

0:0648

0:825

1
1:0926
q1
qI 2 2:0538  0:825
1:5507:
I22 2
1:0926
q1

I12

Once again we notice that the offered formulas of recalculation are especially
21
convenient, when the xed values m21
1 ; m2 are used for any values of initial
currents.

7.1.2

Active Three-Port

We continue Sect. 6.2. Let a subsequent regime correspond to a point M 2 with


2
2
2
; YL2
; YL3
and currents I12 ; I22 ; I32 of loads.
conductivities YL1

7.1 Recalculation of Currents for the Case of Load Changes

211

The non-uniform coordinates


m21

2
YL1
;
2  Y G1
YL1
L1

m22

2
YL2
;
2  Y G2
YL2
L2

m23

2
YL3
:
2  Y G3
YL3
L3

The regime change


2
1
m21
1 m1  m1 ;

2
1
m21
2 m2  m2 ;

2
1
m21
3 m3  m3

7:4

We dene the coordinates of the point M 2 similarly to (6.48). The distance


d24



1 I12
I22
I32

1 :
l4 I1G1 I2G2 I3G2

Then
q n21

I12
;
I1SC

q n22

I22
;
I2SC

q n23

I32
;
I3SC

q n24

d24
:
dSC
4

Repeating a mental step for the plane, we obtain the required transformation
similarly to (7.2) and matrix
2
6
6
J21  6
4

m21
1
0
0

0
m21
2
0

m21
1 1
I1G1

m21
2 1
I2G2

0
0
m21
3

m21
3 1
I3G3

3
0
07
7
:
07
5
1

7:5

From here, we pass to the required currents


I12

qI12

q1

I22

qI22

q1

I32

qI32

q1

I11 m21
1
I11
I1G1

m21
1  1

I21
I2G2

m21
1  1

I21
I2G2

m21
1  1

I21
I2G2

m21
2  1

I31
I3G3

m21
3  1 1

I31
I3G3

m21
3  1 1

I31
I3G3

m21
3  1 1

I21 m21
2
I11
I1G1

m21
2  1
I31 m21
3

I11
I1G1

m21
2  1

;
;
:

These relationships are directly generalized for any number of loads.

7:6

212

7 Recalculation of Load Currents of Active Multi-ports

Special case
Only the load conductivity YL3 is changed. The matrix
2

1
60
6
J21  6 0
4

0
1
0

0
0
m21
3

m21
3 1
I3G3

3
0
07
7
:
07
5
1

Example 2 We continue Example 3 of Sect. 6.2 and rewrite the initial regime
parameters
1
1
1
YL1
0:5; YL2
0:5; YL3
1;

I11 0:974; I21 0:82; I31 1:61;


m11 0:25; m12 0:316; m13 0:387:
The subsequent regime parameters
2
2
2
1; YL2
2; YL3
20;
YL1

I12 1:254; I22 1:356; I32 3:1:


The non-uniform coordinates
m21 0:4;

m22 0:6486;

m23 0:9266:

Regime change (7.4)


m21
1 0:4  0:25 1:6;

m21
2 0:6486  0:3158 2:0538;

m21
3 0:9266  0:387 2:3948:
Matrix (7.5)
2

1:6
6 0
21
J  6
4 0
0:04

0
0
2:0538
0
0
2:394
0:0648 0:0939

3
0
07
7:
05
1

We check up subsequent currents (7.6)


1:6  0:974
1:5592

1:254;
0:04  0:974 0:0648  0:82 0:09396  1:61 1 1:2433
2:0538  0:82
2:3394  1:61
I22
1:356; I32
3:1:
1:2433
1:2433

I12

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

7.2

213

Recalculation of Currents for the Case of Changes


of Circuit Parameters

We consider the concrete circuit (see Sect. 6.1.3) in Fig. 7.2.


The above results allow investigating the influence of a lateral conductivity yN
and longitudinal conductivity y0N on load currents [2]

7.2.1

Change of Lateral Conductivity

Let the conductivity value yN be changed, yN ! yN . Corresponding changes of an


 and short circuit point SC ! SC are shown in Fig. 7.3. On the
initial point M ! M
other hand, the coordinates of the points G1 ; G2 by (6.26) and (6.24) do not depend
on the value yN . Therefore, a straight line G1 G2 is the xed line. The point 0, as the
open circuit regime, does not depend on this element too.
According to Fig. 7.1, the influence of this conductivity yN can be interpreted as
a projective transformation in the plane I1 ; I2 . This transformation, as similar to
I1

y1N

1.25

y0N
V0
5

0.833

YL1
V2

y2

I2

0.25

y2N

V1

y1

0.25

yN

YL2

Fig. 7.2 Example of a circuit with variable conductivities of voltage source


Fig. 7.3 Regime change at
the expense of conductivity
yN ! yN

I2
G2

1
SC

m2(yN)

SC

yN
0

yN yN
m1(yN) 1

G1

yN

yN

m1(yN) 1

G1

yN

I1
yN
m1(yN)

214

7 Recalculation of Load Currents of Active Multi-ports

(7.2) and (7.3), can recalculate the initial currents I1 ; I2 of the point M to the

subsequent currents I1 ; I2 of the point M.
The point M has the homogeneous coordinates n1 ; n2 ; n3 , which are set by
reference triangle G1 0 G2 and a unit point SC. On the other hand, the xed reference triangle G1 0 G2 and a new unit point SC form a subsequent system of
 has the same homogeneous coorprojective coordinates. Therefore, the point M



dinates n1 ; n2 ; n3 ; that is,

qn n:
Using (6.36), we may write
  I;

qn C

qn C  I:

The desired transformation is the multiplication of matrices


 1  C  I J  I;
qI C
 1 has view (6.37).
the matrix C
Therefore, the resultant matrix
2

3
0
0 5;
J33

J11
J 4 0
J31

0
J22
J32

I1SC
;
I1SC

I2SC
;
I2SC

7:7

where its elements


J11
J31
J32

J22

J33

!
SC SC
dSC

I
d
SC3 G1 1SC 3SC  1 ;
I1 d3
d3 I1
!
dSC
I2SC dSC
3
3
SC G2 SC SC  1 :
I2 d3
d3 I2

dSC
3
;
dSC
3

Dividing all these elements by J33 , we obtain the resultant matrix in the form
2

I SC dSC
3
1
I1SC 
dSC
3

6
6
6
J 6 0
6 I1SC dSC
4 I SC d3SC 1
1

I1G1

3
0
I SC dSC
3
2
I2SC dSC
3

I SC dSC
2 3 1
I SC dSC
2 3
I2G2

7
7
0 7:
7
7
5
1

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

215

In this case, the factor q contains the value J33 .


We may introduce transformation parameters
m1 yN

I1SC dSC
3
;
I1SC dSC
3

m2 yN

I2SC dSC
3
:
I2SC dSC
3

7:8

Because the conductivity yN has an equal effect on the load currents, these
parameters are equal to each other; that is, m1 yN m2 yN mN .
Finally, we obtain the above desired transformation
2

3 2
mN
qI1
4 qI2 5 4 0
mN 1
q1
I1G1

0
mN

mN 1
I2G2

3 2 3
0
I1
0 5  4 I2 5 :
1
1

7:9

From here, we pass to the required currents


I1

I1
I1G1

I1 mN
;
mN  1 IIG22 mN  1 1
2

I2 mN
I2
:
I1
I2
m

1

mN  1 1
N
I G1
I G2
1

7:10

The obtained relationships carry out the recalculation of currents at the respective change of conductivity yN in the form of parameter mN . Transformation (7.9) is
a projective transformation and possesses group properties.
Let us formulate the value mN by the initial and subsequent currents I11 ; I11 . To do
this, we use expression (7.10) for the current I1 as I2 0.
Then
I1

I1 m N
I1
I1G1

mN  1 1

7:11

From here, we obtain expression


I1  0
I1  0

:
mN 
G1
I1  I1G1
I1  I1
It is possible to consider this expression as the cross-ratio for points I1 ; I1 relatively to the base points I1 0; I1 I1G1 ; that is,

mN 0 I1

I1


I1G1 :

216

7 Recalculation of Load Currents of Active Multi-ports


mN

I1

mN

SC

I1

I1

SC

I1

G1

I1

I1

Fig. 7.4 Change of the current I1 ! I1 at the change of yN

The points I1 0; I1 I1G1 are xed points and do not depend on the value
yN . Therefore, expression (7.11) is the transformation I1 ! I1 with parameter mN as
it is shown in Fig. 7.4.
Therefore, the cross-ratio has the same value for the points I1SC ; I1SC
mN 0 I1 I1 I1G1 0 I1SC I1SC I1G1
I1  0
I SC  0
I1  0
I SC  0


SC1
 SC1
:
G1
G1
G1
I1  I1
I1  I1
I1  I1
I1  I1G1

7:12

Now, let us formulate the value mN by yN ; yN and determine the sense of the
parameter mN .
Viewing expressions (6.28), we may consider expression (7.8) of mN as a
non-uniform coordinate
mN

I1SC dSC
I SC I SC
3
1SC  1SC n13  n13 :
SC SC
d
I 1 d3
d3
3

Hence, the value mN is the cross-ratio of the initial SC and subsequent points SC
shown in Fig. 7.3.
0
The base values of conductivity yN are values y0N ; yG1
N . The value yN 1
G1
G1
determines the current I1 0. The value yN presets the current I1 I1 as I2 0.
Analysis of the circuit gives the following relationship for yG1
N
yG1
N y0N

y1 y1N
y2 y2N

yiN :
y1 y1N
y2 y2N

7:13

The sense of value yiN will be explained below.


Therefore, the cross-ratio has the view
mN 1 yN

yN

yiN

yN  1 yN  1 yN yiN


:
yN yiN yN yiN yN yiN

7:14

Now, we formulate the value mN by the yN ; yN for the general case of currents
I1 ; I2 .

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

217

Using (6.21), we get SC currents


I1SC V0 Y10 V0 y0N y1N ;
yR

I1SC V0 Y10 V0 y0N

y1N
:
yR

Substituting these currents in (7.8), we obtain


I1SC yR ySER yN SER

;y
y0N y1N y2N ;
I1SC yR ySER yN
 SC
 SC


I1
I2SC
dSC
I1
I2SC
3
G1 G2  1

1
I1
I2
I1G1 I2G2
dSC
3

ySER yN yiN yN



:
ySER yN yiN yN

Finally, we have
mN

I1SC dSC
yN yiN
3

:
SC
SC
yN yiN
I1 d3

This expression is equal to (7.14).


So, the value mN can be interpreted as a relative change of the conductivity yN .
In turn, the value yiN is a scale for yN . The correspondence of the values mN ; yN is
shown in Fig. 7.5.
i
i
Let us clarify the sense of the values yiN and yG1
N yN . The value yN is the
internal conductance of circuit relatively to terminals of disconnected conductivity
yN for the open circuit of loads, as it is shown in Fig. 7.6a.
In turn, if yN yiN , the determinate of matrix Y parameters (6.21), DY 0. In
1
this case, we have the current I1 I1G1 for the rst given load YL1
and the disconnected second load, as it is shown in Fig. 7.6b. And vice versa, the current
1
and the disconnected rst load.
I2 I2G2 is preset for the second given load YL2
Example 3 We use the date of Example 1 for the initial regime
YL1 0:5; YL2 0:5; yN 0:625; I1 0:979; I2 0:8247;
I1SC 3; I2SC 2; l3 dSC
3 0:677:
i

yN

yN

yN

yN

mN

mN

Fig. 7.5 Correspondence of the conductivity yN and transformation parameter mN

218

7 Recalculation of Load Currents of Active Multi-ports

Fig. 7.6 a Internal


conductance yiN of circuit.
b Conductance yiN presets
current I1G1

y1N

(a)
y0N

1.25

y1

y2N

0.25

0.833

2.5

y2

yN

y1N

(b)
y0N
V0
5

0.25

1.25

y1

y2N

0.25

0.833

2.5
+

yN

y2
0.25

Let the subsequent regime parameters be given as


yN 1:25; I11 0:8465; I21 0:713;
I1SC 2:679; I2SC 1:785; l3 dSC
3 0:7116:
Then, transformation parameter (7.8)
mN 0:8494:
Transformation (7.9)
2

3 2
qI1
0:8494
6  7 4
0
4 qI2 5
0:01
q1

0
0:8494
0:009262

3 2I 3
1
0
6 7
5
0  4 I2 5 :
1
1

Subsequent currents (7.10)


0:8494  0:979
0:8315

0:8465;
0:01  0:979  0:009262  0:8247 1 0:9825
I2 0:8494  0:8247 0:713:
0:9825

I1

YL1
G1

I1=I1

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

219

Internal conductivity (7.13)


yiN 2:5

0:25  1:25
0:25  0:833

2:9 :
0:25 1:25 0:25 0:833

We check transformation parameter (7.14)


mN

7.2.2

yN yiN 0:625 2:9


0:8494 :

yN yiN
1:25 2:9

Change of Longitudinal Conductivity

We consider again the circuit in Fig. 7.2. Let the conductivity y0N be changed,
 and short circuit point
y0N ! y0N . Corresponding changes of an initial point M ! M
SC ! SC are shown in Fig. 7.7. Also, the coordinates of points G1 ; G2 by
(6.24) and (6.26) are proportional to the value y0N . Therefore, the subsequent straight
 2 is parallel to the initial line G1 G2 ; the values Y G1 ; Y G2 do not change.
1 G
line G
L1
L2
Let us determine xed points and lines as y0N change. Naturally, the point 0 does
not depend on this element. If the current across conductivity y0N is equal to zero, a
straight line S1 S2 is the xed line and we have the following equation of this line:
y1N
y2N
I1
I2 yi0N  V0 0:
y1 y1N
y2 y2N

Fig. 7.7 Regime change at


the expense of conductivity
y0N ! y0N

I2
G2
G2
SC
SC
M
SC

d3

S1

M
0

S1

I1

G1

S2

I2

G1

S2

I1

220

7 Recalculation of Load Currents of Active Multi-ports

Fig. 7.8 a Internal


conductivity yi0N of circuit.
b Conductivity yi0N presets
the current I1S1

y1N

(a)

1.25

y1

y2N

0.25

N
i

yN

y 0N

0.833

y2
0.25

0.625

y1N

(b)
i

y0N
V0
5

1.25

y1

y2N

0.25

0.833
+

yN
0.625

y2

YL1
S1

I1=I1

0.25

The internal conductivity yi0N of the circuit relatively to terminals of conductivity


y0N , by Fig. 7.8a, has the value
yi0N yN

y1 y1N
y2 y2N

:
y1 y1N
y2 y2N

7:15

The normalized view of the line S1 S2 equation


I1
I2

 1 0;
I1S1 I2S2
where the points


y1
I1S1 yi0N 1
V0 ;
y1N



y2
I2S2 yi0N 1
V0
y2N

7:16

In turn, if y0N yi0N , the determinant DY 0. In this case, we have the load
1
current I1 I1S1 for the rst given load YL1
and the disconnected second load, as it is
shown in Fig. 7.8b. And vice versa, the current I2 I2S2 is preset for the second
1
given load YL2
and the disconnected rst load.

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

221

Therefore, the influence of conductivity y0N can be interpreted as a projective


transformation in the plane I1 ; I2 . In this case, we have the reference triangle S1 0 S2
and a unit point SC, which form the initial system of projective coordinates.
Similarly to (6.36), we may determine the projective coordinates of the point M 1 .
The homogeneous coordinates
q n1

I1
;
I1SC

q n2

I2
;
I2SC

q n3

d3
:
dSC
3

The distances d3 ; dSC


3 to the straight line S1 S2 have the view


1 I1
I2


1
;
l3 I1S1 I2S2


1 I1SC
I2SC


1
:
l3 I1S1
I2S2

d3
dSC
3

7:17

In turn, the non-uniform coordinates


m1

qn1
;
qn3

m2

qn2
:
qn3

For explanation, these non-uniform coordinates are represented in Fig. 7.9.


Fig. 7.9 View of the
non-uniform coordinates

I2
1

m1
0

m1

1 SC
1
1

m2

m2

0
0

S1

I1

S2

222

7 Recalculation of Load Currents of Active Multi-ports

Transformation matrices (6.36) and (6.37)


2
6
6
C 6
4

1
I2SC

1
I1S1 l3 dSC
3

1
I2S2 l3 dSC
3

 l 1dSC

C1

1
I1SC

I1SC
6
0
6
4 SC
I1
I1S1

0
I2SC

0
0

I2SC
I2S2

l3 dSC
3

7
7
7;
5

3 3

3
7
7:
5

On the other hand, the same reference triangle S1 0S2 and a new unit point SC
form the other system of projective coordinates and distance dSC
3 .
Then, the required transformation is similar to (7.9)
3 2 3
3 2
2
I1
qI1
0
0
m 1 y0N
7 6 7
6  7 6
0
m
y

0
2
0N
q
I


5 4 I2 5;
4 25 4
m 1 y0N 1
m 2 y0N 1
1
S1
S2
I
I
q1
1
1

where transformation parameters


m1 y0N

I1SC dSC
3
;
I1SC dSC
3

m2 y0N

I2SC dSC
3
;
I2SC dSC
3

7:18

Because the conductivity y0N has an equal effect on the load currents, these
parameters are equal to each other; that is, m1 y0N m2 y0N m0N :
Finally, we obtain
2
3 2
3 2I 3
qI1
0
0
m0N
1
6  7 4 0
7
m0N 0 5  6
7:19
4 qI2 5
4 I2 5 :
m0N 1 m0N 1
1
S1
S2
I
I
q1
1
1

From here, we pass to the required subsequent currents


I1
I2

I1
I1S1
I1
I1S1

m0N

I1 m0N
;
 1 IIS22 m0N  1 1

m0N

I2 m0N
:
 1 IIS22 m0N  1 1

7:20

Let us formulate the value m0N by the initial and subsequent currents I1 ; I1 .

7.2 Recalculation of Currents for the Case of Changes of Circuit Parameters

223

To do this, we use (7.20) for the current I1 as I2 0. Then, we get


transformation
I1

I1
I1S1

I1 m0N
:
m0N  1 1

7:21

The xed points of this transformation are the points I1 0; I1 I1S1 in


Fig. 7.10.
Similarly to (7.12), it is possible to constitute the cross-ratio

m0N 0 I1

I1


I1S1 0 I1SC


I1S1 0 I1G1

I1SC

I1G1


I1S1 :

Now, we formulate the value m0N by y0N , y0N and determine the sense of this
parameter m0N . According to the above case, we get
m0N 0 y0N y0N  yi0N

y0N  0
y0N  0

:
y0N yi0N y0N yi0N

7:22

The value m0N can be considered as a relative change of the conductivity y0N .
In turn, the value yi0N is a scale for y0N . The correspondence of values m0N and
y0N is shown in Fig. 7.11.
Example 4 We use the date of Example 1 for the initial regime
y0N 2:5; I1 0:979; I2 0:8247;
I1SC 3; I2SC 2; l3 dSC
3 0:677:
Let the subsequent regime parameters be given as
y0N 2; I1 0:896; I2 0:7545;
I1SC 2:6548; I2SC 1:77; l3 dSC
3 1:6969:

m0N

m0N

S1

I1

I1

I1

SC

m0N

SC

G1

G1

I1

I1

y0N

y0N

y0N

m0N

I1

I1

Fig. 7.10 Change of current at change of y0N


0
0

y0N
m0N

Fig. 7.11 Correspondence of the conductivity y0N and parameter m0N

I1

224

7 Recalculation of Load Currents of Active Multi-ports

Transformation parameter (7.18)


m0N 0:9323:
Transformation (7.9)
2

3 2
qI1
0:9323
0
6  7 4
0
0:9323
4 qI2 5
0:011 0:01016
q1

3 2I 3
1
0
6 7
0 5  4 I2 5:
1
1

Subsequent currents (7.20)


0:9323  0:979
0:9127

0:896;
0:011  0:979 0:01016  0:8247 1 1:01914
0:9323  0:8247
0:7688
I2

0:7545:
0:011  0:979 0:01016  0:8247 1 1:01914

I1

Internal conductivity (7.14)


yi0N yN

y1 y1N
y2 y2N

0:625 0:4 1:025:


y1 y1N
y2 y2N

We check transformation parameter (7.22)


y0N
y0N

y0N yi0N y0N yi0N
2
2:5
0:6611


0:9323:

2 1:025 2:5 1:025 0:7092

m0N

7.3

Comparison of Regimes and Parameters


of Active Two-Ports

Let two active two-ports be given with different parameters. Bunches of load
straight lines and the conformity of points of characteristic and running regimes are
presented in Fig. 7.12.
The problem is to nd such values of conductivities and currents of loads of the
 that its relative regimes would be equal to regimes of
second two-port (a point M)
the rst two-port (a point M with preset values of conductivities YL1 ; YL2 and
currents I1 ; I2 ).
For this purpose, we consider systems of projective coordinates of these
 10 G
2
two-ports [1]. Such systems are set by the triangles of reference G1 0 G2 , G
and unit points SC, SC respectively.

7.3 Comparison of Regimes and Parameters

225

Fig. 7.12 Conformity of the


characteristics of two active
two-ports

I2
G2

G2
SC
SC

M
M

G1
G1

I1

Therefore, we have the equality of non-uniform coordinates by (6.28) for the


load conductivities
m1

YL1
YL1

;
G1
G1
YL1  YL1
YL1  YL1

m2

YL2
Y
 L2 G2 :
G2
YL2  YL2
YL2  YL2

7:23

m2
:
m2  1

7:24

From here, we get the formulas


G1
YL1 YL1

m1
;
m1  1

G2
YL2 YL2

Next, it is necessary to nd a transformation which provides the recalculation of


currents.
The conditions of equal regimes lead to the equality of homogeneous projective
coordinates; that is,

qn n:
According to (6.36) for these two-ports, we get
  I:
qn C  I; 
qn C
Therefore, we obtain the necessary transformation as
 1  C  I J  I;

qI C

7:25

226

7 Recalculation of Load Currents of Active Multi-ports

where the matrixes


2

 1
C

I1SC
6 0
4

2
6
C 6
4

0
I2SC

I SC
1
I G1
1

I SC
2
I G2
2

1
I1SC

0
1
I1G1 l3 dSC
3

3
0
0 7
5;

l dSC

7:26

3 3

1
I2SC
1
I2G2 l3 dSC
3

0
 l 1dSC

3
7
7:
5

7:27

3 3

Then, the matrixJ has the form


2

J11
J 4 0
J31

0
J22
J33

3
0
0 5:
J33

7:28

In turn, the matrix elements are


I1SC
I SC
 dSC
l
; J22 2SC ; J33 3 3SC ;
SC
I1
I2
l3 d3
!
I1SC I1G1 l3 dSC
3 dSC
l
3
3
G1

1 ;
I1SCI1G1 l
3 dSC
I1 l3 dSC
3
3
!
I2SC I2G2 l3 dSC
3 dSC
l
3
3
G2

1 :
I2SCI2G2 l
3 dSC
I2 l3 dSC
3
3

J11
J31
J32

The form of these expressions shows that we may divide all the elements by J33
and introduce values
1
m

I1SC l3 dSC
3
;
I1SC l
3 dSC
3

2
m

I2SC l3 dSC
3
:
I2SC l
3 dSC
3

7:29

These values state a quantitative estimation of the circuit distinctions.


Then, the required transformation becomes
3 2
qI1
6  7 6
4 qI2 5 4
2

q1

1
m
 G1 0

1 I1

m

1
G1
G1
1

I
I
1

0

m
 G2 2

1 I2

m

1
G2
G2
2

I
I
2

3 2 3
I1
0
6 7
07
5  4 I2 5:
1
1

7:30

7.3 Comparison of Regimes and Parameters

227

In turn,
I1
I2

I1
I1G1

I1
I1G1

 G1
I1
I G1
1

 G1
I1
I G1
1

1
I1 m

1  1
m


I2
I2G2

2
I2 m

1  1
m

I2
I2G2

 G2
I2
I G2
2

 G2
I2
I G2
2


;
2  1 1
m


2  1 1
m

7:31

The structure of expressions (7.31) shows that it is possible to introduce the


normalized values
I G1
 N1 m
 1 1G1 ;
m
I1

I G2
 N2 m
 2 2G2 :
m
I2

Then, we obtain the normalized form of transformation (7.30)


2


q

6
4q


I1
I G1
1
I2
I G2
2

1
q

6
7 6
56
6
4

 N1
m
0
 N1  1
m

2I 3
1
7
I1G1
7
6
I2 7
 N2
m
07
7  4 I2G2 5:
5
1
N
2  1 1
m
0

7:32

The obtained transformations allow carrying out the recalculation of currents of


the second two-port for any currents of the rst two-port, using the preset values
 1; m
 2 or m
 N1 ; m
 N2 :
m
Example 5 We consider two circuits similar to the circuit in Fig. 6.8 and use some
date of example 2 of Sect. 6.1.3.
Let the following different conductivities be given for these two-ports as
y1N 1:25 ! y1N 0:8928;
y2N 0:8333 ! y1N 0:5681:
The bunch centers of the second two-port
G1
I1G1 16; YL1
1:1428;
G2
G2
I2 18; YL2 0:8181:

Let the relative regimes of these two-ports be equal to each other by (7.23)
m1 0:25;

m2 0:3158:

For the second circuit we nd the following values.

228

7 Recalculation of Load Currents of Active Multi-ports

Load conductivities (7.24)


m1
0:3809;
m1  1
m2
G2
0:3776:
YL2
m2  1

G1
YL1 YL1

YL2
SC currents

I1SC 2:434;

I2SC 1:549:

The distances
3 
dSC
l
3 0:7618;

3 d13 0:9211;
l

3 0:08362:
l

Distinctions (7.29)
 1 0:721;
m

 2 0:6882:
m

Transformation (7.30)
3 2
 I1
q
0:721
6  7 4
 I2 5
0
4q
0:0216
1
q
2

3 2I 3
1
0
0
6 7
0:6882 0 5  4 I2 5:
0:0233 0
1

Currents (7.31)
0:721  0:979
0:7058

0:7355;
0:0216  0:979  0:0233  0:8247 1 0:9596
I2 0:6882  0:8247 0:5915:
0:9596

I1

7.4

Comparison of Regime of Active Two-Ports


with Linear Stabilizations of Load Voltages

We use the circuit in Fig. 6.20. For convenience, we redraw this circuit in Fig. 7.13.
Let us consider the two similar circuits with different values of element parameters and regime parameters. It is necessary to prove an approach to the comparison
or recalculation of running regimes of such circuits. The characteristics of these
comparable circuits are given in Fig. 7.14. The condition of regime comparison is the
conformity of characteristic regimes as a projective transformation [4].

7.4 Comparison of Regime of Active Two-Ports

229

y1N
y0N

V1

y2N

V2

V0
+

yN

YL1

YL2

Fig. 7.13 Active two-port with stabilization of load voltages

Fig. 7.14 Conformity of


characteristics of similar
circuits

y2N

I2

SC

y2N

I0M

I0M
G2

y2N

G2

G2

G2

y2N
1

y2N
SC

y2N

M
M
0

SC

I1

I0M
1

y1N

y1N

G1 G1

I0M

SC

y1N

y1N

Case of equal regimes


 1 ) will be
The running regimes of these circuits (a point M1 corresponds to a point M
equivalent or equal to each other if their non-uniform coordinates (6.79) and (6.80)
are equal, and homogeneous projective coordinates (6.82) are proportional; that is,
 1 ; m2 m
 2 ; q n n:
m1 m

7:33

Using relationships (6.82) and (6.83), we nd the required conformity of load


currents
 1  n C
 1  C  I M  I:
q I C

7:34

230

7 Recalculation of Load Currents of Active Multi-ports

The resulting matrix M has the view


2

M11
M 4 0
M31

0
M22
M32

3
0
0 5;
M33

7:35

I1SC
I SC
 dSC
l
; M22 2SC ; M33 3 3SC ;
SC
I1
I2
l3 d3
SC

I1SC 1
I SC 1
 d
 dSC
l
l
G1 SC  G13 3 SC ; M32 2G2 SC  G23 3 SC :
I I
I I
I ld
I l d

M11
M31

3 3

3 3

The similar condition for the conformity of regulator conductivities follows from
(6.79) and (6.80). In particular, the conductivity y1N is dened by the equality

y11N

y1
ySC
y11N
ySC
 SC 1N G1 1 1N G1  SC 1N G1 :
G1
 y1N y1N  y1N y1N  y1N y1N  y1N

7:36

Case of not equal regimes


 2 . The point M2
Let a subsequent regime of the second circuit be given by a point M

of equal regime of the rst circuit corresponds to this point M2 .
Therefore, it is possible to consider the points M1 , M2 of the rst circuit as the
change of its regime; that is, M1 ! M2 . Then, the same change of regime will be for
1 ! M
 2 . Therefore, this regime change denes the difference
the second circuit, M
 2:
of regimes of comparable circuits, M1 ! M
Let us validate an expression of regime change via the change of non-uniform
 21 of subsequent regime, the
projective coordinates. The non-uniform coordinate m
 2 , is given by (6.79)
point M
 21 0 y21N ySC
m
yG1
1N 
1N

y21N
ySC
1N

:
y21N  yG1
ySC
yG1
1N
1N  
1N

The regime change are naturally expressed through the cross-ratio


 21
m
y21N y11N yG1
1 0 
1N

y21N
y11N
 21  m
 11 :

m
y21N  yG1
y11N  yG1
1N
1N

This change is also equal to the change of regime of the rst circuit
21
2
1
 21
m
1 m1 m1  m1 :

7:37

7.4 Comparison of Regime of Active Two-Ports

231

Let us express these changes through load currents. Using (6.81) and (6.31), we
obtain
m21
1

I12 l3 d13
;
I11 l3 d23

m21
2

I22 l3 d13
:
I21 l3 d23

7:38

21

Let the changes m21
1 ; m2 and the initial equal regime, the points M1 ; M1 , be
given.
At rst, it is necessary to nd load currents of these regime changes, the points
 2 . Similarly to (7.2), we have for the second circuit
M2 ; M

i
h i
2
1
qI J21  I ;

7:39

The matrix
2

m21
1
21
6
0
J 4
1
21
I G1 m1  1

0
m21
2
1
21
I G2 m2  1

3
0
07
5:
1

From here, we pass to the required currents


I12
I22

I 1
1
I G1

I11 m21
1
I 1

21
2
m21
1  1 I G2 m2  1 1

I 1
1
I G1
1

21
2
m21
1  1 I G2 m2  1 1

;
7:40

I21 m21
2
I 1

Similar relationships are obtained for the rst circuit.


At last, we nd the required expression of recalculation for comparable circuit
 2:
currents, the conformity M1 ! M
Using (7.39) and (7.34), we obtain
2
1
21
qI  J21   I  J21   M  I1  J   I1 :

7:41

The resultant matrix J  has the view


21

J11
21
J  4 0
J31

0
J22
J32

3
0
0 5;
J33

7:42

232

7 Recalculation of Load Currents of Active Multi-ports

where
I1SC
SC
3 dSC
l
21 I2
3
J11 m21


;
J

m
;
J

;
22
33
1 SC
2 SC
I1
I2
l3 dSC
3
I1SC 1
3 dSC
l
3
J31 m21
;
1 G1 SC  G1
I1 I1
I l dSC
1

3 3

3 3

I2SC 1
3 dSC
l
3
J32 m21

:
2 G2 SC
I2 I2
I G2 l dSC
From the obtained expressions, the procedure of regime comparison of two
circuits follows. For this purpose, we consider an example.
Example We use the date of Example 5 for Sect. 6.3. The element parameters have
the following values:
V0 5 ; y0N 2:5 ; yN 0:625:
We have for the rst circuit.
The characteristic values of currents and conductivities
SC
SC
I1G1 3:125; yG1
1N 3:125; y1N 6:25; I1 1:785;
SC
SC
I2G2 6:25; yG2
2N 6:25; y2N 4:166; I2 5:357:

The currents of the initial regime, point M1


I11 1; I21 1:
The conductivities of the regulators
y11N 0:735;

y12N 2:777:

The non-uniform projective coordinates


m11 0:461;

m12 0:154:

The distances of the points M1 ,SC to the straight line G1 G2


l3 d13 0:52;

l3 dSC
3 0:428:

We have for the second circuit.


The element parameters have the same values, except yN 0:25:

7.4 Comparison of Regime of Active Two-Ports

233

The characteristic values of currents and conductivities


I1G1 4:25; yG1
SC
ySC
1N 4:25; 
1N 6:25; I1 2:309;
I2G2 7; yG2
SC
ySC
2N 7; 
2N 4:166; I2 5:706:
2
 1G
The distance of point SC to straight line G
3 dSC
l
3 0:358:
Case of equal regimes
Elements of matrix (7.35)
2:309
5:706
1:293; M22
1:0652;
1:785
5:357
2:309 1
0:358

0:0365;

4:25 1:785 3:125  0:428


5:706 1
0:358

0:0183:

7 5:357 6:25  0:428

M11
M31
M32

M33

0:358
0:836;
0:428

Currents (7.34)
M11 I11
M32 I21 M33
1:293  1
1:293

1:4528;

0:0365  1 0:0183  1 0:836


0:89
M22 I21
I21
M31 I11 M32 I21 M33
1:0652  1
1:0652

1:194:

0:0365  1 0:0183  1 0:836


0:89
I11

M31 I11

Conductivities of regulators (6.76)


y y
2:75
y11N I11 G2 0N 1 N 1 1:45
0:916:
I2  I1 I2
7  2:646
y y
2:75
y12N I21 G1 0N 1 N 1 1:194
2:047:
I1  I1 I2
4:25  2:646
Let us check up the equality of non-uniform coordinates (7.33).

234

7 Recalculation of Load Currents of Active Multi-ports

Non-uniform coordinates (7.36)


0:916
6:25

0:274  0:595 0:461 m11 ;
0:916  4:25 6:25  4:25
2:0475
4:166
 12

0:226  1:47 0:154 m12 :
m
2:0475 7 4:166 7
 11
m

Case of not equal regimes


Let us consider the second circuit. We believe that the regime corresponds to the
 2 . Let the currents be given as
point M
I12 1:5;

I22 2;

Then, the conductivities


y21N 1:179;

y22N 7:333:

Regime change (7.37)


m21
1

1:179
0:916

0:3839  0:2749 1:396:
1:179  4:25 0:916  4:25

Similarly, we get
7:333
2:0475
m21
2 7:333 7  2:0475 7 0:5116  0:2263 2:261:

The currents of the second circuit by (7.40)


1:45  1:397
2:02

1:5;
1:45  0:0935 1:194  0:18 1 1:35
I22 1:194  2:261 2:
1:35

I12

Matrix (7.42)
2

1:8073
J  4 0
0:1574
21

0
2:4089
0:2103

3
0
0 5:
0:8368

 2 by (7.41)
We check the currents of point M
1:8073  1
1:8073

1:5;
0:1574  1 0:2103  1 0:836 1:2045
I22 2:4089 2:
1:2045

I12

References

235

References
1. Penin, A.: Normalized representation of the equations of active multiport networks on the basis
of projective geometry. Moldavian J. Phys. Sci. 10(34), 350357 (2011). http://sfm.asm.md/
moldphys/2011/vol10/n3-4/index.html Accessed 30 Nov 2014
2. Penin, A.: Recalculating the load currents of an active multiport with variable parameters on the
basis of projective geometry. Electrichestvo 10, 6673 (2012)
3. Penin, A.: Recalculation of the loads current of active multiport networks on the basis of
projective geometry. J. Circ. Syst. Comput. 22(05), 1350031, 13 p. (2013). doi: 10.1142/
S021812661350031X http://www.worldscientic.com/doi/abs/10.1142/S021812661350031X
Accessed 30 Nov 2014
4. Penin, A.: Comparison of regimes of active two-port networks with stabilization of load
voltages. Int. J. Electron. Commun. Electr. Eng. 3(6), 118 (2013). https://sites.google.com/site/
ijeceejournal/

Chapter 8

Passive Multi-port Circuits

8.1

Input-Output Conformity of Four-Ports as an Afne


Transformation

We use the results of Sect. 6.1.1. Let us consider a four-port circuit with changeable
load voltage sources in Fig. 8.1 and give necessary relationships I3 V1 ; V2 ; I4 V1 ; V2
for this network.
Taking into account the specied directions of currents, this network is described
by the following system of Y parameters equations
2

I1

Y11
6I 7 6
6 2 7 6 Y12
6 74
Y13
4 I3 5
Y14
I

Y12
Y22
Y23
Y24

Y13
Y23
Y33
Y34

3 2 V1 3
Y14
6V 7
Y24 7
27
76
6 7 Y  V;
Y34 5 4 V3 5
Y44
V

8:1

From here, we obtain


"

I3

I4

 " # " SC;SC #


I3
V1
Y23

SC;SC ;
Y24
V2
I4

Y13

Y14

8:2

where SC currents of both loads


"

I3SC;SC
I4SC;SC

Y33

Y34

 " #
V3
Y34

:
Y44
V4

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_8

8:3

237

238

8 Passive Multi-port Circuits


I3

V3

I1

V1

Y
+

I2

I4

+
_

V2

V4
+

Fig. 8.1 Four-port with changeable load voltage sources V1 ; V2

The inverse expression of (8.2)


"

V1
V2


1 Y
34 24
DY Y14

#
 " SC;SC
 I3
I3
Y23
 SC;SC
;
Y13
I4
 I4

8:4

where determinant D34


Y Y13 Y24  Y23 Y14 .
Therefore, we get
I4  I4SC;SC

Y24
D34
I3  I3SC;SC Y V1 ;
Y23
Y23

8:5

I4  I4SC;SC

Y14
D34
I3  I3SC;SC  Y V2 :
Y13
Y13

8:6

Equation (8.5) determines a family of parallel load straight lines with a


parameter V1 . Similarly, Eq. (8.6) determines a family of parallel load straight lines
with parameter V2 . These load straight lines are shown in Fig. 8.2 for characteristic
regimes (as the short circuit V1S C 0; V2S C 0 and open circuit V1OC ; V2OC ) and a
running regime V11 ; V21 .
We consider that the load currents dene the rectangular Cartesian system of
coordinates I3 0I I4 . Then, the load voltages correspond to the system of afne
coordinates V1 0V V2 .
The equation of the axis V1 is dened by the expression (8.6) as V2 0
I4  I4SC;SC

Y14
I3  I3SC;SC :
Y13

8:7

This straight line passes through the point SC; SC or the point 0V . The value
Y14 =Y13 corresponds to a slope angle. The currents I3SC;SC ; I4SC;SC determine the point
SC; SC.
Similarly, the equation of the axis V2 is dened by the expression (8.5) as V1 0

8.1 Input-Output Conformity of Four-Ports


Fig. 8.2 Systems of
Cartesian coordinates I3 0I I4
and afne coordinates
V1 0V V2

239
I4
1,SC

I3

SC,SC

I4

(SC, SC)

0V

1,SC

I4

V1
OC,SC

I4

V1

V1

SC,1

I4

I4

OC

V2

1
1

I3

SC,OC

I4

OC

V2

OC,OC

I4

(OC, OC)

V2
OC,SC

SC,SC

I3
0I

I4  I4SC;SC

OC,OC

I3

Y24
I3  I3SC;SC :
Y23

I3
SC,OC

I3

I3

SC,1

I3

8:8

The open circuit regime of both loads corresponds to the point OC; OC or to
point 0I . Let us determine the voltages V1OC ; V2OC . From (8.1), we get
"

I1

I2

Y11

Y12

 " # " SC;SC #


I1
V1
Y12
SC;SC ;

Y22
V2
I2

8:9

where SC currents of both loads


"

I1SC;SC

 " #
V3
Y14

:
Y24
V4

8:10

#
 " SC;SC
 I1
I1
Y12
 SC;SC
;
Y11
I2
 I2

8:11

Y13

Y23

I2SC;SC
The inverse expression of (8.9)
"

V1
V2


1 Y22
12
DY Y12

2
where determinant D12
Y Y11 Y22  Y12 .
Using (8.11), we get, similarly to (6.7) and (6.8), that

V1OC

Y22 SC;SC Y12 SC;SC


I1
12 I2
;
D12
DY
Y

8:12

240

8 Passive Multi-port Circuits

V2OC

Y12 SC;SC Y11 SC;SC


I1
12 I2
:
D12
DY
Y

8:13

We determine now the currents I3OC;OC ; I4OC;OC by (8.2)


I3OC;OC Y13 V1OC  Y23 V2OC I3SC;SC ;

8:14

I4OC;OC Y14 V1OC  Y24 V2OC I4SC;SC :

8:15

We must map the point OC; OC onto the axes V1 ; V2 by parallel lines to these
axes. Then, the coordinates or components I3OC;SC ; I4OC;SC will correspond to the
point V1OC on the axis V1 . Using (8.12) and (8.2) as V2 0, we get
I3OC;SC Y13 V1OC I3SC;SC ;

8:16

I4OC;SC Y14 V1OC I4SC;SC :

8:17

In turn, the coordinates I3SC;OC ; I4SC;OC correspond to the point V2OC on the axis V2 .
Using (8.13) and (8.2) as V1 0, we obtain
I3SC;OC Y23 V2OC I3SC;SC ;

8:18

I4SC;OC Y24 V2OC I4SC;SC :

8:19

Let an initial regime be given by values V11 ; V21 or by a point M 1 . Then, the
coordinates I31;SC ; I41;SC dene the point V11 on the axis V1 . Using (8.2) as V2 0, we
have
I31;SC Y13 V11 I3SC;SC ;

8:20

I41;SC Y14 V11 I4SC;SC :

8:21

In turn, the coordinates I3SC;1 , I4SC;1 dene the point V21 on the axis V2 . Using (8.2)
as V1 0, we have
I1SC;1 Y23 V21 I3SC;SC ;

8:22

I4SC;1 Y24 V21 I4SC;SC :

8:23

Let us introduce normalized coordinates for the point M 1 . The point OC; OC is
a length scale for the system of coordinates V1 0V V2 .
Then, similarly to (6.17)(6.20), the normalized coordinates have the view

8.1 Input-Output Conformity of Four-Ports

n11

241

V11  0V
V11
I31;SC  I3SC;SC

;
V1OC  0V V1OC I3OC;SC  I3SC;SC

8:24

V21
I4SC;1  I4SC;SC

:
V2OC I4SC;OC  I4SC;SC

8:25

n12

These normalized coordinates, as an afne ratio, are the invariants of afne


transformation (8.2).
Similarly to (2.7), we get the other afne ratios in the form
N11

N21

V11
I31;SC  I3SC;SC

;
 V11 I3OC;SC  I31;SC

8:26

V21
I4SC;1  I4SC;SC
SC;OC
:
1
 V2 I4
 I4SC;1

8:27

V1OC

V2OC

The presented approach will be used for transmission of two measuring signals.
Example 1 We consider the following four-port circuit with given parameters in
Fig. 8.3.
System of Eq. (8.1)
2

I1

0:6813
6I 7 6
6 2 7 6 0:1393
6 74
0:147
4 I3 5
0:0464
I

0:1393
0:7727
0:087
0:159

0:147
0:087
0:3247
0:029

3 2 V1 3
0:0464
6V 7
0:159 7
27
76
6 7:
0:029 5 4 V3 5
0:2803
V
4

Currents (8.3)
"

I3SC;SC

0:3247

0:029

I4SC;SC

y3K

12

I3

V4
10

0.25

I4

I2

V2

0.333

V1

y10

y2L

0.666

0.2

y4L

yKL

yK0

I1

y1K

0.416

V3

#
 " # "
12
3:6064
0:029


:
0:2803
10
2:455

yL0

y20

0.2

0.25

Fig. 8.3 Four-port represents two two-port networks connected by conductivity yKL

242

8 Passive Multi-port Circuits

System of Eq. (8.2)


"

I3
I4

0:147

0:0464

#
 " # "
V1
3:6064
0:087


:
0:159
2:455
V2

The inverse expression


"

V1
V2


1 0:159
34
DY 0:0464

#
 "
3:6064  I3
0:087
;

0:147
2:455  I4

D34
Y 0:147  0:159  0:087  0:0464 0:01934:
Equation (8.7) of the axis V1
I4  2:455 0:3156  I3  3:6064:
Equation (8.8) of the axis V2
I4  2:455 1:8275  I3  3:6064:
The obtained system of coordinates is shown in Fig. 8.4.
SC currents (8.10) of both loads
"

I1SC;SC

I2SC;SC

0:147

0:087

#
 " # "
12
2:228
0:0464


:
0:159
10
2:634

The voltages V1OC ; V2OC by (8.12) and (8.13)


0:7727
0:1393
2:228
2:634 4:119;
0:507
0:507
0:1393
0:6813
2:228
2:634 4:1514;

0:507
0:507
0:6813  0:7727  0:13932 0:507:

V1OC
V2OC
D12
Y

Fig. 8.4 Example of the


Cartesian coordinates I3 0I I4
and afne coordinates
V1 0V V2

I4

0V

2.5

OC

V1 =4.119

V1=2.202

V2=2.663

V1

1.5

M
(OC, OC)

1
OC

V2 =4.1514

V2

1
2

2.5

3.5

I3

8.1 Input-Output Conformity of Four-Ports

243

The currents I3OC;OC ; I4OC;OC by (8.14) and (8.15)


I3OC;OC 0:147  4:119  0:087  4:1514 3:6064 2:6397;
I4OC;OC 0:0464  4:119  0:159  4:1514 2:455 1:6038:
The components I3OC;SC ; I4OC;SC by (8.16) and (8.17)
I3OC;SC 0:147  4:119 3:6064 3:0;
I4OC;SC 0:0464  4:119 2:455 2:263:
The components I3SC;OC ; I4SC;OC by (8.18) and (8.19)
I3SC;OC 0:087  4:1514 3:6064 3:2452
I4SC;OC 0:159  4:1514 2:455 1:795:
Let the initial regime, the point M 1 , be given by voltages
V11 2:202;

V21 2:663:

Currents (8.2)
"

I31
I41

0:147

0:0464

# "
# "
#
 "
2:202
3:6064
3:051
0:087


:
0:159
2:663
2:455
1:93

The coordinates I31;SC ; I41;SC by (8.20) and (8.21)


I31;SC 0:147  2:202 3:6064 3:2827;
I41;SC 0:0464  2:202 2:455 2:3528:
The coordinates I3SC;1 ; I4SC;1 by (8.22) and (8.23)
I1SC;1 0:087  2:663 3:6064 3:3747;
I4SC;1 0:159  2:663 2:455 2:032:
Normalized coordinates (8.24) and (8.25)
2:202 3:2827  3:6064

0:5346;
4:119
3:0  3:6064
2:663
2:032  2:455
n12

0:6415:
4:1514 1:795  2:455
n11

244

8 Passive Multi-port Circuits

Afne ratios (8.26) and (8.27)


2:202
3:2827  3:6064

1:1446;
4:119  2:202
3:0  3:2827
2:663
2:032  2:455
N21

1:789:
4:1514  2:663 1:795  2:032

N11

8.2

Input-Output Conformity of Four-Ports


as a Projective Transformation

Now, we consider a four-port circuit in Fig. 8.5. Let us give necessary relationships
between input currents and load conductivities [1]. This network is described by
system of Eqs. (8.1). Also, we have that
V1 I1 =YL1 ;

8.2.1

V2 I2 =YL2 :

Output of a Four-Port

The above circuit concerning loads represents an active two-port network.


Therefore, we use the results of Sect. 6.1.3. As it was shown in Fig. 6.9, the family
of straight lines I1 ; I2 ; YL1 0; I1 ; I2 ; YL2 0 at change of YL1 ; YL2 is represented by two bunches of straight lines in the system of coordinates I1 0 I2 . For
convenience, we show this family in Fig. 8.6.
Next, we use the idea of projective coordinates of a running regime point. Let the
initial or running regime corresponds to the point M 1 , which is set by the values of
1
1
; YL2
and currents I11 ; I21 .
the conductivities YL1

I3

I1

V3

V1

I4

V4

I2

YL1
V2

YL2

Fig. 8.5 Four-port with changeable load conductivities YL1 ; YL2

8.2 Input-Output Conformity of Four-Ports


Fig. 8.6 Two bunches of
load straight lines with the
parameters YL1 ; YL2

245
YL2

I2
G2

G2

YL2

G2

I2

YL2=

SC

SC

SC

I2

SC

YL2
1

I2

1
SC

G1

G1

SC

I1

I1

I1
1

YL1

I1
G1

YL1=

YL1

YL1

m 1= 1

m 1=

m1

m1

In addition, this point is dened by projective non-uniform coordinates m11 ; m12


and homogeneous coordinates n11 ; n12 ; n13 which are set by the reference triangle
G1 0G2 and a unit point SC.
We rewrite requirement relationships. Non-uniform projective coordinate (6.28)
and (6.29)
m11

1
YL1
;
1  Y G1
YL1
L1

m12

1
YL2
:
1  Y G2
YL2
L2

8:28

In turn, homogeneous projective coordinates n1 ; n2 ; n3 set the non-uniform


coordinates by (6.30)
m1

qn1
;
qn3

m2

qn2
;
qn3

8:29

where q is a proportionality factor. Homogeneous coordinates (6.36) are dened by


load currents
2

qn1

1
7
6
7
6
I SC
7 61
6
6 qn2 7 6 0
7 4
6
7
6
1
5
4
I1G1 l3 dSC
3
qn3

where l3 dSC
3

 SC

I1
I1G1

I2SC
I2G2

0
1
I2SC

1
I2G2 l3 dSC
3


1 .

I1

I1

6 7
6 7
6 7
6 7
6 7
7 6 7
0 7  6 I2 7 C  6 I2 7;
6 7
5 6 7
6 7
6 7
1
SC
5
4
4 5
l3 d3
1
1
0

8:30

246

8 Passive Multi-port Circuits

From here, the non-uniform coordinates assume the convenient form


m1
m2

I1
I1G1 l3 dSC
3

1
I
I1SC 1
I G2 lI2 dSC
3 3
2

 l 1dSC

I1
I1G1 l3 dSC
3

1
I
I2SC 2
I G2 lI2 dSC
3 3
2

 l 1dSC

;
8:31

3 3

3 3

The inverse transformation of (8.30)


2 3
2
3
3
2 SC
2 3
qI1
n1
I1
0
0
n1
6
7
6 qI 7
SC
6 2 7 60
7 6 n2 7
7
1 6
I
0
2

C



6
7 4 SC
5 6 7
4 n2 5 ;
I1
I2SC
4 5
4
5
SC
l3 d3
n3
I1G1
I2G2
q1
n3

8:32

Therefore, the currents


I1

qI1
I1SC m1
I SC
;
I2SC
SC
q1
1
m

m

l
d
G1
G2
1
2
3
3
I
I
1

I2 I SC

1
I1G1

8.2.2

m1

8:33

I2SC m2
I2SC
I2G2

m2  l3 dSC
3

Input of a Four-Port

Let us consider the input currents I3 ; I4 of our four-port. We may superpose the
system of coordinates I3 0 I4 with the system of coordinates I1 0 I2 in Fig. 8.7.
Then, any point with coordinates I1 ; I2 corresponds to a point with coordinates
I3 ; I4 . In terms of geometry, a projective transformation takes place which
transfers points of the plane I1 ; I2 into points of the plane I3 ; I4 . Therefore, the
reference triangle G1 0 G2 , point SC, and running regime point M 1 correspond to
1

the triangle G1 0 G2 , point SC, and point M , as it is shown by arrows in Fig. 8.7.
Next, the axes of currents I1 ; I2 correspond to the axes I 1 ; I 2 . Also, two bunches
of the straight lines I1 ; I2 ; YL1 0, I1 ; I2 ; YL2 0 correspond to two bunches of
the lines I3 ; I4 ; YL1 0, I3 ; I4 ; YL2 0 with centers in the points G2 ; G1 . Thus,
1

the point M is set by other currents I31 ; I41 . Also, this point is dened by projective
non-uniform and homogeneous coordinates which are set by the reference triangle
G1 0 G2 and a unit point SC.
1

The property of projective transformations shows that the point M coordinates


1
are equal to the point M 1 coordinates, as these points M 1 , M are set by the same

8.2 Input-Output Conformity of Four-Ports


Fig. 8.7 Projective
transformation of the plane
I1 ; I2 onto plane I3 ; I4

247
I2

I4

G2

I2

G2

YL2

YL2

I1

G1

YL1

0
0

Fig. 8.8 Distances of points


1
M , SC to the axis I 1

I3
G1

YL1

I4

I1

I1

G1

I4

SC

SC

I4

SC

I4

G1

OC

I4

OC

I3

SC

1
I3 I 3

G1

I3

I3

1
1
loads YL1
; YL2
. Therefore, this property gives required invariant relations between
input and output currents.
1
For nding of the point M projective coordinates, it is necessary to obtain
equations of sides of reference triangle. The normalized equation of the side 0 G1
or axis I 1 in Fig. 8.8 looks like

I4OC

I4
k 1 I3

 1 0;
 k 1 I3OC I4OC  k1 I3OC

k 1 tga1

I4G1  I4OC
;
I3G1  I3OC

where k1 is an angular coefcient or slope ratio.


1

Then, the point M distance d2 to the axis I 1 is dened by expression


I41
k1 I 1
 OC 3 OC  1;
OC
 k 1 I3
I4  k 1 I3
v
!2
!2
u
u
1
k1
t
l2
OC
;
I4OC  k1 I3OC
I4  k1 I3OC
1

l2 d2

I4OC

248

8 Passive Multi-port Circuits

where l2 is a normalizing factor.


SC

The point SC distance d2 to the axis I 1 is


SC

l 2 d2

I4SC
k1 I3SC

 1:
I4OC  k 1 I3OC I4OC  k 1 I3OC

Similarly, the axis I 2 equation is


I4
k 2 I3

1 0;
k2 I3OC  I4OC k 2 I3OC  I4OC
k2 tga2

I4G2  I4OC
:
I3G2  I3OC

Then, the point M distance d1 to the axis I 2 is


I41
k2 I 1
 OC 3 OC 1;
OC
 I4
k 2 I3  I4
v
!2
!2
u
u
1
k2
t
l1

:
k2 I3OC  I4OC
k2 I3OC  I4OC
1

l1 d1

k 2 I3OC

SC

The point SC distance d1 to the axis I 2 is


SC

l 1 d1

I4SC
k2 I3SC

1:
k2 I3OC  I4OC k2 I3OC  I4OC

Similarly, the innitely remote straight line 1 equation is

I4G1

I4
k 1 I3
G1
 1 0;
G1
k 1 I3
I4 k1 I3G1

k1

I4G2  I4G1
:
I3G1  I3G2

The point M distance d3 to the line 1 is


I41
k 1 I31
G1
 1;
G1
k 1 I3
I4 k1 I3G1
v
!2
!2
u
u
1
k1
l3 t G1
G1
:
I4 k1 I3G1
I4 k1 I3G1
1

l 3 d3

I4G1

8.2 Input-Output Conformity of Four-Ports

249

SC

The point SC distance d3 to the line 1 is


SC

l3 d3

I4SC
k1 I3SC

 1:
I4G1 k 1 I3G1 I4G1 k 1 I3G1

The homogeneous projective coordinates are


1

qn11

d1

SC
d1

k2
k2 I3OC
1

k 2 I3OC 

I1
SC 4
OC
I4 l1 d1

qn12

d2
SC
d2

I4OC

k1

d3

SC

I1
SC 3

 k1 I3OC l2 d2

SC

l1 d1

I4OC

I1
SC 4
OC
k1 I3 l2 d2

qn13

I1
SC 3

 I4OC l1 d1

1
l2 d2

k1

I1
SC 3
I4G1 k1 I3G1 l3 d3
1
1

I1 
:
SC 4
SC
G1
G1
I4 k 1 I3 l3 d3
l 3 d3
SC
d3

The matrix form of these expressions


2

qn1

I3

C11
7
6 7
6
7
6 7 6
6
7
6 7 6
6
6 qn2 7 C  6 I4 7 6 C21
7
6 7 6
6
7
6 7 4
6
5
4 5
4
C 31
qn3
1

C 11
k2
C 21
k1
C 31
k1

1
SC
l1 d 1
 1SC
l2 d 2
 1SC
l3 d 3

I3

6 7
7 6 7
7 6 7
7  6 I4 7:
7 6 7
5 6 7
4 5
1

The constituents of this matrix


C 11
C 31

k2
k2 I3OC 

SC
I4OC l1 d1

k1
SC

I4G1 k1 I3G1 l3 d3

;
:

C 21

k1
SC

I4OC  k1 I3OC l2 d2

8:34

250

8 Passive Multi-port Circuits

From here, the non-uniform coordinates have the form similar to (8.31)

m1

m2

C 11 I3
C 31 I3

C31
I
k1 4

C 21 I3
C 31 I3

C11
I
k2 4

C21
I
k1 4

C 31
I
k1 4

1
SC
l3 d 3

1
SC
l1 d 1

1
SC
l2 d 2

1
SC
l3 d3

;
8:35
:

The obtained expressions have the general appearance in comparison with (8.31)
because of the non-orthogonal coordinates I 1 0I 2 .
In practice, characteristic values of input current, output currents (as vertexes of
coordinate triangles), and loads are precalculated or preprogrammed by the calculation or testing of four-port network.
Further, using running values of input currents, we nd or, more precisely,
restore values of non-uniform coordinates (8.35) and values of given load conductivities according to the expressions YL1 m1 ; YL2 m2 . These expressions are
inverse to expression (8.28).
Such a formulated algorithm represents a practical interest for transfer of two
sensing signals via an unstable four-port network or a three-wire line; it is analogous to a signal transmission via a two-port network of Chap. 4.
The inverse transformation of (8.34)
2

qI3

n1

7
7
6
6
7
7
6
6
7
7
6
6
6 qI4 7 C1  6 n2 7
7
7
6
6
7
7
6
6
5
5
4
4
q1
n1

1
C11
6 1
6 C11
4
1
C11

I4G1
I3G1
1
I3G1

1
C 12
1
C 12
1
C 12

I4G2
I3G2
1
I3G2

1
C13
1
C13
1
C13

1

1

k 2 I3OC  I4OC
SC
I G1 l1 d1 ;
OC
G1 3
k 1  k2 I3  I3

C 12 
1

C 13

n1

6 7
6 7
7 6 7
7  6 n2 7 :
5 6 7
6 7
1
4 5
OC
I3
n3
I4OC
I3OC

The constituents of this matrix


C 11

k 1 I3OC  I4OC
SC
I G2 l2 d2 ;
k 1  k2 I3OC  I3G2 3

k 1 I3G1 I4G1
SC
I OC l3 d3 :
OC
G1 3
k 1 k 1 I3  I3

8:36

8.2 Input-Output Conformity of Four-Ports

251

Similarly to (8.33), we may pass to the currents


1

I3

I4

1

1

C11 m1 C12 m2 C 13

C 11

1 1
I3G1

m1 C12

1 1
I3G2

m2 C 13

1 I4G1
I3G1
1 1
C 11 I G1
3

m1 C12

1 I4G2
I3G2
1 1
m1 C12 I G2
3

m2 C 13

C 11

1 1
I3OC

1 I4OC
I3OC
1 1
m2 C 13 I OC
3

;
8:37
:

The obtained expressions have the general view in comparison with (8.33)
because of non-orthogonal coordinates.
Convenience of expressions (8.35) and (8.37) consists in their identical form; we
replace load conductivities by their non-uniform projective coordinates and currents
are already non-uniform coordinates too.

8.2.3

Recalculation of Currents at Load Changes

2
2
Let a subsequent regime be corresponded to a point M 2 with loads YL1
; YL2
,
2
2
2 2
2 2
non-uniform coordinates m1 ; m2 , output currents I1 ; I2 , and input currents I3 ; I4 .
At rst, using the result of Sect. 7.1.1, we rewrite subsequent currents I12 ; I22 (7.2)

qI12

I11

I11

6 27
6 7
6 7
1
4 qI2 5 J21   4 I21 5 C21   C  4 I21 5:
q1

The matrix
2

21
J11
J21  4 0
21
J31

0
21
J22
21
J32

3 2 m21
1
0
60
5

0
4 21
m1 1
1
I G1
1

0
m21
2

m21
2 1
I2G2

In turn, the matrix


C21 1 C1  M21 ;
where
2

m21
1
21
4
M  0
0

0
m21
2
0

3
0
0 5:
1

3
0
07
5:
1

8:38

252

8 Passive Multi-port Circuits

Here, we use the regime changes


2
1
m21
1 m1  m1 ;

2
1
m21
2 m2  m2 :

8:39

These changes of the non-uniform coordinates are also true for the input currents. Therefore, it is possible at once to obtain similar relationships for the
recalculation of the input currents.
2

qI32

I31

I31

21 6
7
6 27
6 7
1
4 qI4 5 C  M21   C  4 I41 5 J   4 I41 5:

q1

8:40

21

We calculate the matrix J . This matrix has the general view in comparison
with (8.38)
2

21

3
21
J 13
21 7
J 23 5:
21
J 33

21

J
21
6 11
J  4 J 21
21
21
J 31

J 12
21
J 22
21
J 32

8:41

Using transformations (8.38) and (8.40), we nd the subsequent currents


I12

1
qI12
m21
1 I1
m21 1
;
21 1
m
q1
1
1
1
2
G1 I
G2 I 1
1

I1

I22

I2

1
qI 2
m21
2 I2
2 m21 1
:
21 1
m
1
1
q1
1
2
G1 I
G2 I 1
1

I1

I2

8:42

21

21

21

21

21

21

I32

qI 2 J I31 J 12 I41 J 13
3 11
;
q1 J 21 I 1 J 21 I 1 J 21
31 3
32 4
33

I42

qI 2 J I31 J 22 I41 J 23
4 21
:
q1 J 21 I 1 J 21 I 1 J 21
31 3
32 4
33

8:43

We note that the denominators of expressions (8.42) and (8.43) are equal to each
other.

8.2.4

Two Cascaded Four-Port Networks

Let us consider cascaded four-port networks in Fig. 8.9. The rst four-port is given
by a matrix Y 36 of Y parameters and the second four-port corresponds to a matrix
Y 14 .

8.2 Input-Output Conformity of Four-Ports

253

I3

I5

I1
V1

V5

I6

3-6

I4

1-4

I2

YL1
V2

V6
+

YL2

Fig. 8.9 Cascaded four-port networks

Fig. 8.10 Mapping of


reference triangles

I2, I4, I6
G2

G2

G2
0
0
0

G1
I1, I3, I5

G1
G1

Similarly, we may superpose the system of coordinates I5 0 I6 with the systems


of coordinates I3 0 I4 , I1 0 I2 in Fig. 8.10.
Then, a projective transformation, which transfers points of the plane I1 ; I2 into
points of the plane I5 ; I6 , takes place. Therefore, the reference triangle G1 0G2
~ 2 . Also, a unit point SC, running regime point M 1
~ 1 ~0G
corresponds to the triangle G
1
~
~
~
correspond to points SC; M . Moreover, two bunches of the straight lines
I1 ; I2 ; YL1 0; I1 ; I2 ; YL2 0 correspond to two bunches of the lines
~ 2; G
~ 1.
I5 ; I6 ; YL1 0; I5 ; I6 ; YL2 0 with centers in the points G
~ 1 is dened by projective non-uniform and homogeneous
Also, the point M
~C.
~
~ 2 and a unit point S
~ 1 ~0G
coordinates which are set by the reference triangle G
1
~ are equal to the projective coordinates
These projective coordinates of the point M
1
1
of the points M ; M according to the property of projective transformations. Thus,
this property gives required invariant relationships between input and output currents of cascaded networks.
~ 1 are obtained similarly to projective
Projective coordinates of the point M
1
coordinates of the point M . For this purpose, it is necessary to form equations of
~ 2.
~ 1 ~0G
sides of the reference triangle G

254

8.2.5

8 Passive Multi-port Circuits

Examples of Calculation

Let the four-port be given in Fig. 8.11.


We use Example 1 and rewrite system of Eqs. (8.1)
I Y  V;
2
0:6813 0:1393
6 0:1393
0:7727
6
Y 6
4 0:147 0:087

0:147

0:0464 0:159

0:0464

0:087
0:3247

0:159 7
7
7:
0:029 5

0:029

0:2803

For the output of the four-port we have the next results


Short circuit currents (8.10)
I1SC 2:228;

I2SC 2:634:

System of Eq. (8.9)


"

I1
I2

2
3
 " #
I1SC;SC
V1
Y12
5
4

Y22
V2
I2SC;SC
#

 " # "
V1
2:228
0:6813 0:1393

:
0:1393
0:7727
V2
2:634


Y11

Y12

Parameters of bunch centers G1 ; G2 (6.26) and (6.24)


Y11 SC;SS
0:6813
2:634 2:228 15:11;
I
I1SC;SS
0:1393
Y12 2
1 SC;SS
2:634
I G1
G1
18:9; YL1

I2

L1
0:8;
G1
Y12
0:1393
VL1
Y22 SC;SS
0:7727
2:228 2:634 15;

I
I2SC;SS
0:1393
Y12 1
1 SC;SS
2:228
I G2
G2
16; YL2

I1

L2
0:9375:
G2
Y12
0:1393
VL2

I1G1
V1G1
I2G2
V2G2

Fig. 8.11 Example of the


four-port

y3K
0.416

V3
12

I3

V4

I4

0.666

y10
0.25

y2L

0.333

10

0.2

y4L

yKL

yK0

I1

y1K

V1

I2
V2

yL0

y20

0.2

0.25

YL2

YL1

8.2 Input-Output Conformity of Four-Ports

255

Let the initial regime be given by


1
YL1
0:5;

I11 1:101;

1
0:333;
YL2

V11 2:202;

I21 0:8868;

V21 2:663:

Non-uniform projective coordinates (8.28)


m11
m12

1
YL1

1
YL1
0:5
0:3846;

G1
0:5 0:8
 YL1

1
YL2
0:333
0:2622:

G2
1
0:333
0:9375
YL2  YL2

Matrix (8.30)
2

1
2:228

C 4 0

1
15:110:6769

1
2:634
1
150:6769

0
0

3
5;

1
0:6769

For this matrix, the value



l3 dSC
3

I1SC
I SC
2G2  1
G1
I1
I2

2:228 2:634

 1 0:6769:
15:11
15

Homogeneous projective coordinates (8.30)


I11
1:101
I21
0:8868
0:4942; qn12 SC
0:3366;

SC
2:228
2:634
I1
I2
1:101
0:8868
1
qn13

15:11  0:6769 15  0:6769 0:6769


0:1076  0:08734 1:4773 1:2823:

qn11

Let us check up the non-uniform projective coordinates


m11

qn1 0:4942
0:3846;

qn3 1:2823

m12

qn2 0:3366
0:2622:

qn3 1:2823

Inverse matrix (8.32)


2

2:228
C1 4 0
0:1474

0
2:634
0:1757

3
0
5:
0
0:6769

256

8 Passive Multi-port Circuits

Currents (8.33)
2:228  0:3846
0:86

1:101;
0:1474  0:3846 0:1757  0:2622 0:6769 0:7797
2:634  0:2622
I21
0:8868:
0:7797

I11

For the input of the four-port we have the next results


The currents, corresponding to the short circuit,
I3SC;SC 3:6064;

I4SC;SC 2:455:

Using (8.2), we may nd the bunch centers G1 ; G2 . Then, we get the following
systems of equations:
2
3
" G1 # 
 " G1 #
I3
I3SC;SC
V1
Y13 Y23
5;

4

SC;SC
Y14 Y24
0
I4G1
I4
" G2 # 
"
# 2 SC;SC 3

0
I3
I3
Y13 Y23
5:


4
Y14 Y24
V2G2
I4G2
I4SC;SC
Therefore,
I3G1 0:147  18:9 3:6064 6:4423;
I4G1 0:0464  18:9 2:455 3:333;
I3G2 0:087  16 3:6064 5;
I4G2 0:159  16 2:455 5:
The currents, corresponding to the open circuit,
I3OC 2:6397;

I4OC 1:6038:

The currents of the point M by (8.2)


I31 3:051;

I41 1:93:

The normalized equation of the axis I 1


k1

I4OC

I4G1  I4OC 3:333  1:6038


0:4617;

I3G1  I3OC 6:442  2:6397

I4
k 1 I3
I4
I3

 1 0:
 OC
1
OC
OC
0:3835 0:8305
 k 1 I3
I4  k 1 I3

8.2 Input-Output Conformity of Four-Ports


1

The point M distance to the axis I 1


1:93
3:051

 1 0:3566;
0:3835
0:8305
s

2 
2
1
1
2:8721:
l2

0:3835
0:8305
1

l2 d2

The point SC distance to the axis I 1


SC

l2 d2

2:455
3:606

 1 1:0576:
0:3835 0:8305

The normalized equation of the axis I 2


I4
k 2 I3
I4
I3

1 0;

1
2:1961 1:5258
k2 I3OC  I4OC k2 I3OC  I4OC
k2

I4G2  I4OC 5  1:6038


1:4392:

I3G2  I3OC 5  2:6397


1

The point M distance to the axis I 2


1:93
3:051

1 0:1216;
2:196
1:5258
s

2 
2
1
1
0:798:
l1

2:1961
1:5258
1

l1 d1

The point SC distance to the axis I 2


SC

l1 d1

2:455
3:606

1 0:2458:
2:1961 1:5258

The normalized equation of the innitely remote line 1


I4
k 1 I3
I4
I3

 1 0;

1
11 9:166
I4G1 k1 I3G1 I4G1 k1 I3G1
k1

I4G2  I4G1 5  3:333


1:2:

I3G1  I3G2 6:442  5

257

258

8 Passive Multi-port Circuits


1

The point M distance to the line 1


1:93 3:051

 1 0:4918;
11
9:166
s
 2 
2
1
1
0:142:
l3

11
9:166
1

l3 d3

The point SC distance to the line 1


SC

l3 d3

2:455 3:606

 1 0:3835:
11
9:166

The homogeneous projective coordinates have the same values


1

qn11

d1
SC
d1

0:1216
0:4942;
0:2458

0:3566
0:3366;
1:057

0:4918
1:2823:
0:3835

qn12

d2
SC
d2
1

qn13

d3
SC
d3

Matrix (8.34)
2

1
0:24581:5258
6
1
4 1:0570:8305
1
 0:38359:166

1
0:24582:1961
1
1:0570:3834
1
 0:383511

2:666
1:852
6
4 1:1383 2:4653
0:2844 0:237

1
0:2458
7
1
1:057 5
1
0:3835

3
4:067
7
0:9454 5:

2:607

Let us carry out the requalication of non-uniform projective coordinate (8.35)


2:666  3:051  1:8522  1:93  4:067
0:495

0:3846:
0:2844  3:051  0:237  1:93 2:607 1:282
1:1383  3:051 2:4653  1:93  0:9454 0:339
m12

0:2622:
0:2844  3:051  0:237  1:93 2:607
1:282

m11

8.2 Input-Output Conformity of Four-Ports

259

The matrix of inverse transformation (8.36)


2

1

C

0:9422
4 0:4915
0:1474

0:8789
0:8791
0:1757

3
1:7868
1:0847 5:
0:6768

We check input currents (8.37)


0:9422  0:3846 0:8789  0:2622 1:7868 2:3796

3:051;
0:1474  0:3846 0:1757  0:2622 0:6768 0:7796
0:4915  0:3846 0:8791  0:2622 1:0847 1:5042
I41

1:93:
0:7796
0:7796

I31

For the recalculation of currents we have the next results


Let the subsequent regime, the point M 2 , be given by
2
1;
YL1

I12 1:459;

V12 1:459;

2
1;
YL2

I22 1:602;

V22 1:602:

Non-uniform projective coordinates (8.28)


m21

1
0:555;
1 0:8

m22 0:516:

Currents (8.2)
I32 3:253;

I42 2:132:

Regime changes (8.39)


2
1
m21
1 m1  m1 0:555  0:3846 1:442;
2
1
m21
2 m2  m2 0:516  0:2622 1:968:

Matrix (8.38)
2

1:442
J21  4 0
0:0292

0
1:968
0:0645

3
0
0 5:
1

We may check up recalculation formulas (8.42) for the output current I1


I12

1:442  1:101
1:5876

1:459:
0:0292  1:101 0:0645  0:8868 1 1:0893

260

8 Passive Multi-port Circuits

Matrix (8.41) for the input current I3


2
1:1463
21
21
J  4
J 21
0:01927

1:3225
21
J 22
0:2982

3
2:503
21
J 23 5:
0:5728

We check up recalculation formula (8.43) for the current I3


I32

8.3

1:1463  3:051 1:3225  1:93  2:503


3:545

3:253:
0:01927  3:051 0:2982  1:93 0:5728 1:0893

Transmission of Two Signals Over Three-Wire Line

We continue with the transfer of signals in Sect. 4.3.1. We consider three-wire


communication line with losses. Therefore, the interference of transmitted signals
takes place. Using the measured input currents of the line, we may restore these two
signals. To do this, let us consider the following two examples.

8.3.1

Transmission by Using of Cross-Ratio

We may use non-uniform coordinates or cross-ratios (8.28) as transferable analog


signals VS1 ; VS2 . The corresponding appliance is shown in Fig. 8.12 [3].
We believe that the communication line is equivalent to the resistive four-port in
Fig. 8.11. Then, for a short time, when parameters of the four-port do not change, ve
pairs of conductivities are connected to the output terminals of the line by turns. This
connection is realized by the contact pairs 1, 2, 3, 4, 5 of two multichannel switches
with a pulse generator Gen. The sets of ve pairs of input and output currents of the
line correspond to these pairs of conductivities, as it is shown in Table 8.1.
So, the non-uniform coordinates m1 ; m2 of the point MYL1 ; YL2 are set by
cross-ratios (8.28) of four points
YL1
;
G1
YL1  YL1
YL2
OC
SC G2
G2
m2 YL2
YL2 YL2
YL2 0 YL2 1YL2

:
G2
YL2  YL2
OC
SC G1
G1
YL1 YL1
YL1 0 YL1 1YL1

m1 YL1

Since these cross-ratios m1 ; m2 are equal to the transferable signals VS1 ; VS2 , then
the information conductivities YL1 ; YL2 are expressed by the formulas
YL1

G1
YL1
VS1
;
VS1  1

YL2

G2
YL2
VS2
:
VS2  1

8:44

8.3 Transmission of Two Signals Over Three-Wire Line

261

Fig. 8.12 System of


transmission signals VS1 ; VS2
by cross ratio

5
G1

YL1

YL1

YL1 =

SC

G1

I1
V0

Line

V1

1
OC

YL1 =0

I2

F
VS1

Gen

I3

I4
5

F
VS1

-1

G2

YL2

YL2 =

YL2

SC

VS2

G2

V2

1
OC

YL2 =0

VS2

Table 8.1 Correspondence between output conductivities, input currents, and output currents
Number of set

Load conductivities

Output currents

Input
currents

YL1 0

2
3

YL1
YL1 1

YL2 0

I1 0

I2 0

YL2
YL2 1

I1

I2

I3OC
I3

I4OC
I4

YL1

G2
YL2

I1SC

I2SC

I3SC

I4SC

I2G2
I2G1

I3G2
I3G1

I4G2

G1
YL1

YL2

I1G2
I1G1

I4G1

The function unit F works according to these expressions. The information


conductivities YL1 ; YL2 may directly present measuring thermistors or resistive-strain
sensors.
In turn, the inverse function unit F 1 calculates the transmitted signals VS1 ; VS2
by formulas (8.35)

VS1 m1

VS2 m2

C 11 I3

C 11
I
k2 4

1
SC
d1

C 31 I3

C 31
I
k1 4

1
SC
d3

C 21 I3

C 21
I
k1 4

1
SC
d2

C 31 I3

C 31
I
k1 4

1
SC
d3

;
8:45
:

262

8 Passive Multi-port Circuits

So, it is possible to separate (or restore) two signals by only the input currents
(ve pairs of current values) of line. We note also that multiplicative and additive
errors of the measurement of input conductivities are mutually reduced in expression (8.45). Therefore, such a communication line is subject to action of electromagnetic noises to a lesser degree.

8.3.2

Transmission by Using of Afne Ratio

Afne ratios (8.24) and (8.25) or (8.26) and (8.27) are accepted as transferable
analog signals VS1 ; VS2 . The corresponding appliance is shown in Fig. 8.13 [2].
Similarly, for a short time, four pairs of voltage sources are connected to the output
terminals of the line by turns. This connection is realized by the contact pairs 1, 2,
3, 4 of two multichannel switches with a pulse generator Gen. The sets of four pairs
of input and output currents of the line correspond to these pairs of voltages, as it is
shown in Table 8.2. The voltages V1SC ; V2SC are minimum values; in particular case,
these values equal zero. The voltages V1OC ; V2OC are maximum values; these values
are greater than the information values V1 ; V2 .
So, the normalized coordinates of the point MV1 ; V2 are set by afne ratios
(8.24) and (8.25)
n1

V1  V1SC
;
V1OC  V1SC

n2

V2  V2SC
:
V2OC  V2SC

Fig. 8.13 System of


transmission signals VS1 ; VS2
by afne ratios

4
3

V1

I1
V0

Line

F
SC

I2

V1

OC

V1

VS1

Gen

I3

I4
4

F
VS1

-1

VS2

2
1

F
SC

V2

OC

V2

VS2

8.3 Transmission of Two Signals Over Three-Wire Line


Table 8.2 Correspondence
between load voltages and
input currents

263

Number of set

Load voltages

Input currents

V1SC

V2SC

I3SC;SC

I4SC;SC

V1OC

V2SC

I3OC;SC

I4OC;SC

V1SC

V2OC

I3SC;OC

V1

V2

I3

I4SC;OC
I4

Since these afne ratios n1 ; n2 are equal to transferable signals VS1 ; VS2 , then the
information voltages V1 ; V2 are expressed by the formulas
V1 V1SC  V1OC  V1SC VS1 ;

V2 V2SC  V2OC  V2SC VS2 :

8:46

The function unit F works according to these expressions.


In turn, the inverse function unit F 1 calculates the transmitted signals VS1 ; VS2
by formulas (8.24) and (8.25)
VS1

I31;SC  I3SC;SC
I3OC;SC

I3SC;SC

VS2

I41;SC  I4SC;SC
I4OC;SC  I4SC;SC

8:47

For other afne ratios (8.26) and (8.27), we get similarly to (8.46) that
V1

V1SC VS1 V1OC


;
1  VS1

V2

V2SC VS2 V2OC


:
1  VS2

8:48

The function unit F works according to these expressions. Also, the inverse
function unit F 1 calculates the transmitted signals VS1 ; VS2 by formulas
VS1

I31;SC  I3SC;SC

;
1;SC

I3OC;SC  I3

VS2

I4SC;1  I4SC;SC
I4SC;OC  I4SC;1

8:49

These two variants of use of afne ratios allow to separate (or restore) two
signals by only the input currents (four pairs of current values) of line. We note also
that multiplicative and additive errors of measurement of input conductivity values
are mutually reduced in expressions (8.47) and (8.49).

8.4

Input-Output Conformity of a Balanced Six-Port

Let us continue with the matter we began in Sect. 6.2.2. We consider the currents
I4 ; I5 ; I6 of the voltage sources V4 ; V5 ; V6 for the balanced network in Fig. 6.18.
Let us superpose the coordinate system I4 ; I5 ; I6 with the coordinate system
I1 ; I2 ; I3 in Fig. 8.14. Then, any point with coordinates I1 ; I2 ; I3 corresponds to a

264

8 Passive Multi-port Circuits

Fig. 8.14 Superposition of


coordinate systems of output
I1 ; I2 ; I3 and input I4 ; I5 ; I6
currents

I3
I6

G3

I3
G3

G2

I 2 , I5

G2

SC
I2

SC

G1

I1

G1

I 1 , I4

point with coordinates I4 ; I5 ; I6 . In particular, the running regime point M 1 , a unit


1
point SC and coordinate tetrahedron 0G1 G2 G3 correspond to the points M ; SC and
0G1 G2 G3 . These tetrahedrons are the generalization of the reference triangles for a
plane in Fig. 8.7.
For convenience, the conformity of the output and input characteristic regime
points is shown in Table 8.3.
Using Fig. 6.19, we determine the input current for the point G1 . According to
(6.61), the currents
I5G1 I5SC ;

I6G1 I6SC :

8:50

In turn, according to (6.63), the current


I4G1 V4 UaG1 y4a :

Table 8.3 Output-input


conformity of the six-port

8:51

Output
Points

I1

I2

I3

Input
Points

I4

I5

I6

I4OC

I5OC

I6OC

G1

I1G1
0

G1

I4G1

I5G1

I6G1

I4G2
I4G3
I4SC

I5G2
I5G3
I5SC

I6G2

G2

I2G2

G2

I3G3
I3SC

G3

G3

SC

I1SC

I2SC

SC

I6G3
I6SC

8.4 Input-Output Conformity

265

For the point G2 , we get similarly


I4G2 I4SC y4a V4 ;

I6G2 I6SC I6G1 ;

I5G2 V5 UbG2 y5b :

8:52

Likewise for the point G3 , the currents


I4G3 I4SC I4G2 ;

I5G3 I5SC I5G1 ;

I6G3 V6 UcG3 y6c :

8:53

It is known that projective coordinates of points M 1 and M , concerning the own


coordinate systems, are equal among themselves. Therefore, we must introduce the
1
known coordinates n11 ; n12 ; n13 ; n14 of the point M .
These homogeneous coordinates are dened as the ratio of the distances
1

SC

SC

SC

SC

d1 ; d2 ; d3 ; d4 for the point M and distances d1 ; d2 ; d3 ; d4 for a unit point SC to


the planes of the coordinate tetrahedron 0G1 G2 G3 .
At rst, we get the equations of these planes.
The plane 0G1 G2 equation has the general view
A12 I4 B12 I5 C12 I6 D12 0:
Matrix determinants
OC
I
5

A12 I5G1

I G2
5
OC
I
4

C12 I4G1

I G2
4

I6OC
I6G1
I6G2
I5OC
I5G1
I5G2


1

1 ;

1

1

1 ;

1


OC OC
I
I4
1
6


B12 I6G1 I4G1 1 ;


I G2 I G2 1
6
4
OC OC OC
I
I5
I6
4

G1
G1
D12 I4
I5
I6G1 :


I G2 I G2 I G2
4
6
5

We reduce A12 ; B12 by expressions (8.50) and (8.52). Then



OC OC
I
I6
1
5


A12 I5G1 I6SC 1 I5G2  I5G1 I6OC  I6SC ;


I G2 I SC 1
6
5

OC OC
I
I4
1
6


B12 I6SC I4G1 1 I4G1  I4G2 I6OC  I6SC :


I SC I G2 1
6

Also, it turns out


I5G2  I5G1

y4a
y5b V5 ;
yab

I4G1  I4G2

y4a
y5b V4 :
yab

8:54

266

8 Passive Multi-port Circuits

If the voltage V4 V5 , then A12 B12 .


The plane 0G2 G3 equation
A23 I4 B23 I5 C23 I6 D23 0:
Matrix determinants

A23

C23

OC
I
5

I5G2

I G3
5
OC
I
4

I4G2

I G3
4

I6OC
I6G2
I6G3
I5OC
I5G2
I5G3


1

1 ;

1

1

1 ;

1


OC OC
I
I4
1
6


B23 I6G2 I4G2 1 ;


I G3 I G3 1
6
4
OC OC OC
I
I5
I6
4

G2
D23 I4
I5G2 I6G2 :


I G3 I G3 I G3
4
6
5

8:55

If V5 V6 , then B23 C23 .


The plane 0G1 G3 equation
A13 I4 B13 I5 C13 I6 D13 0:
Matrix determinants

A13

C13

OC
I
5

I5G1

I G3
5
OC
I
4

I4G1

I G3
4

I6OC
I6G1
I6G3
I5OC
I5G1
I5G3


1

1 ;

1

1

1 ;

1


OC OC
I
I4
1
6


B13 I6G1 I4G1 1 ;


I G3 I G3 1
6
4
OC OC OC
I
I5
I6
4

G1
D13 I4
I5G1 I6G1 :


I G3 I G3 I G3
4
6
5

8:56

If V4 V6 , then A13 C13 .


For the plane G1 G2 G3 , we have
A1 I4 B1 I5 C1 I6 D1 0;


G1 G1
G1 OG1
I
I
I6
1
I4
1
6
5




A1 I5G2 I6G2 1 ; B1 I6G2 I4G2 1 ;




I G3 I G3 1
I G3 I G3 1
6
5
6
4

G1 G1 G1
G1 G1
I
I
I5
1
I5
I6
4
4

G2 G2 G2
G2 G2
C1 I4
I5
1 ; D1 I4
I5
I6 :




I G3 I G3 I G3
I G3 I G3 1
4
4
6
5
5

8:57

8.4 Input-Output Conformity

267

If V4 V5 V6 , then A1 B1 C1 .
At last, we give expressions for the distances.
1

SC

The distances d1 ; d1 correspond to the plane 0G2 G3 , therefore


1

l1 d1 A23 I41 B23 I51 C23 I61 D23 ;


SC

l1 d1 A23 I4SC B23 I5SC C23 I6SC D23 :


1

SC

The distances d2 ; d2 correspond to 0G1 G3 , then


1

l2 d2 A13 I41 B13 I51 C13 I61 D13 ;


SC

l2 d2 A13 I4SC  B13 I5SC C13 I6SC D13 :


1

SC

The distances d3 ; d3 match to 0G1 G2 , in that case


1

l3 d3 A12 I41 B12 I51 C12 I61 D12 ;


SC

l3 d3 A12 I4SC B12 I5SC C12 I6SC D12 :


1

SC

Similarly, d4 ; d4 correspond to G1 G2 G3 and


1

l4 d4 A1 I41 B1 I51 C1 I61 D1 ;


SC

l4 d4 A1 I4SC B1 I5SC C1 I6SC D1 :


Finally, the homogeneous coordinates n11 ; n12 ; n13 ; n14 have the view
1

qn11

l 1 d1
SC
l 1 d1

A23

I1
SC 4

l 1 d1

qn12

l 2 d2
SC
l 2 d2

A13

I1
SC 4

l 2 d2

qn13

qn14

l 3 d3
SC
l 3 d3

A23 I41 B23 I51 C23 I61 D23


A23 I4SC B23 I5SC C23 I6SC D23

B23

I1
SC 5

l1 d1

C23

I1
SC 6

l 1 d1

D23
SC

l1 d1

A13 I41 B13 I51 C13 I61 D13


A13 I4SC B13 I5SC C13 I6SC D13

B13

I1
SC 5

l2 d2

C13

I1
SC 6

l 2 d2

D13
SC

l2 d2

A12 I41 B12 I51 C12 I61 D12


A12 I4SC B12 I5SC C12 I6SC D12

A12

B12 1
C12 1
D12
I1
I
I
;
SC 4
SC 5
SC 6
SC
l 3 d3
l3 d3
l 3 d3
l3 d3
1
l 4 d4
A1 I41 B1 I51 C1 I61 D1

SC
A1 I4SC B1 I5SC C1 I6SC D1
l 4 d4
A1

I1
SC 4
l 4 d4

B1

I1
SC 5
l4 d4

C1

I1
SC 6
l 4 d4

D1
SC

l4 d4

268

8 Passive Multi-port Circuits

The conformable matrix form


2

qn11

I41

6 17
6 17
6I 7
6 qn2 7
6
7 C  6 5 7;
6 17
6 17
4 I6 5
4 qn3 5
1
qn4
2 A
23
6 l dSC
6 1 1
6 A
6 13
6
6 l dSC
6
C 6 2 2
6 A12
6
6 l dSC
6 3 3
6
4 A1
SC

l 4 d4

B23

C23

SC
l1 d1

SC
l1 d1

B13

C13

SC
l2 d2

B12
SC

l3 d3
B1

SC

l4 d4

SC

l2 d2
C12

SC

l3 d3
C1

SC

l4 d4

D23 3
SC
l1 d1 7
7
D13 7
7
7
SC 7
l2 d2 7
7:
D12 7
7
SC
l3 d3 7
7
7
D1 5

8:58

SC

l4 d4

In turn, the non-uniform coordinates have the view


SC

m11

qn11 l4 d4 A23 I41 B23 I51 C23 I61 D23

;
A1 I41 B1 I51 C1 I61 D1
qn14 l1 dSC
1

m12

qn12 l4 d4 A13 I41 B13 I51 C13 I61 D13

;
A1 I41 B1 I51 C1 I61 D1
qn14 l2 dSC
2

SC

8:59

SC

m13

qn1 l d A12 I41 B12 I51 C12 I61 D12


31 4 4SC
:
qn4 l3 d3 A1 I41 B1 I51 C1 I61 D1

So, we use only the input currents for (8.59). In practice, the characteristic values
are obtained by calculation or testing the multi-port only by manipulations at the
output terminals. It is technically more convenient than determination of all the Y
parameter.
In turn, we obtain or restore the load conductivities
1
YL1

G1 1
YL1
m1
;
1
m1  1

YL2

G2 1
YL2
m2
;
1
m2  1

YL3

G3 1
YL3
m3
:
1
m3  1

8:60

8.4 Input-Output Conformity

269

Table 8.4 Parameters of the coordinate tetrahedron


Regime points

Input currents
I4

I5

I6

I4OC 0:1793

I5OC 0:2986

I6OC 0:1382

G1

I4G1 2:6

I5G1 2:0

I6G1 0:6666

G2

I4G2
I4G3
I4SC

3:6

I6G2 0:6666

2:0

I6G3 2:2666

1:501

I6SC 0:5786

G3
SC

1:0

I5G2
I5G3
I5SC

1:0
0:8272

Example 2 We continue Example 4 of Sect. 6.2.2. The calculated values of the


characteristic regime points are shown in Table 8.4.
Determinates (8.54)


0:2986 0:1382 1




A12 2:0
0:6666 1 0:8454;


3:6
0:6666 1


0:1382 0:1793 1




B12 0:6666
2:6
1 0:8454;


0:6666
2:6
1


0:1793 0:2986 1




C12 2:6
2:0
1 6:5954;


1:0
3:6
1


0:1793 0:2986 0:1382




D12 2:6
2:0
0:6666 0:5079:


1:0
3:6
0:6666
The plane 0G1 G2 equation
0:8454I4  0:8454I5 6:5954I6  0:5079 0:
Determinates (8.55)

A23

B23


0:2986


3:6

2:0

0:1382


0:6666

2:2666

0:1382
0:6666
2:2666
0:1793
1:0
1:0


1

1 6:1264;

1

1

1 1:3131;

1

270

8 Passive Multi-port Circuits

C23

D23


0:1793


1:0

1:0

0:1793


1:0

1:0


0:2986 1

3:6
1 1:3131;

2:0
1

0:2986 0:1382

3:6
0:6666 0:5253:

2:0
2:2666

The plane 0G2 G3 equation


6:1264I4  1:3131I5  1:3131I6  0:5253 0:
Determinates (8.56)

A13

B13

C13

D13


0:2986


2:0

2:0

0:1382


0:6666

2:2666

0:1793


2:6

1:0

0:1793


2:6

1:0

0:1382
0:6666
2:2666
0:1793
2:6
1:0
0:2986
2:0
2:0
0:2986
2:0
2:0


1

1 2:7222;

1

1

1 4:7186;

1

1

1 2:7222;

1

0:1382

0:6666 0:5382:

2:2666

The plane 0G1 G3 equation


2:7222I4  4:7186I5 2:7222I6 0:5382 0:
Determinates (8.57)

A1


2:0


3:6

2:0


0:6666 1

0:6666 1 2:56;

2:2666 1

B1


0:6666


0:6666

2:2666


2:6 1

1 1 2:56;

1 1

8.4 Input-Output Conformity

C1

D1

271


2:6


1

1

2:6


1

1

2:0
3:6
2:0
2:0
3:6
2:0


1

1 2:56;

1

0:6666

0:6666 13:4825:

2:2666

The plane G1 G2 G3 equation


2:56I4 2:56I5 2:56I6  13:4825 0:
We rewrite the values of the initial regime
1
0:2;
YL1

1
YL2
0:25;

1
YL3
0:1666;

I11 0:4232;

I21 0:5673;

I31 0:335;

I41 0:4921;

I51 0:8654;

I61 0:354:

SC

The distances d1 ; d1

l1 d1 6:1264  0:4921  1:3131  0:8654


 1:3131  0:354  0:5253 0:8883;
SC
l1 d1

6:1264  0:8272  1:3131  1:501


 1:3131  0:5786  0:5253 1:8147:

SC

The distances d2 ; d2
1

l2 d2 2:7222  0:4921  4:7186  0:8654


2:7222  0:354 0:5382 1:242;
SC
l2 d2

2:7222  0:8272  4:7186  1:501


2:7222  0:5786 0:5382 2:7176:

SC

The distances d3 ; d3

l3 d3 0:8454  0:4921  0:8454  0:8654


6:5954  0:354  0:5079 0:6792;
SC

l3 d3 0:8454  0:8272  0:8454  1:501


6:5954  0:5786  0:5079 1:3399:

272

8 Passive Multi-port Circuits


1

SC

The distances d4 ; d4
1

l4 d4 2:56  0:4921 2:56  0:8654


2:56  0:354  13:4825 9:101;
SC

l4 d4 2:56  0:8272 2:56  1:501


2:56  0:5786  13:4825 6:0411 :
The homogeneous coordinates
1

qn11

l 1 d1
SC
l1 d1

0:8883
0:4895
1:8147

6:1264 1 1:3131 1 1:3131 1 0:5253


I 
I 
I 
1:8147 4 1:8147 5 1:8147 6 1:8147
3:376I41  0:7236I51  0:7236I61  0:2895;

qn12

l 2 d2
SC
l2 d2

1:242
0:457
2:7176

2:7222 1 4:7186 1 2:7222 1 0:5382


I
I 
I 
2:7176 4 2:7176 5 2:7176 6 2:7176
1:0017I41 1:7363I51  1:0017I61  0:198;

qn13

l 3 d3
SC
l3 d3

0:6792
0:5069
1:3399

0:8454 1 0:8454 1 6:5954 1 0:5079


I 
I
I 
1:3399 4 1:3399 5 1:3399 6 1:3399
0:6309I41  0:6309I51 4:9I61  0:379;

qn14

l 4 d4
SC
l4 d4

9:101
1:5065
6:0411

2:56 1
2:56 1
2:56 1 13:4825
I 
I 
I
6:0411 4 6:0411 5 6:0411 6
6:0411
1
1
1
0:4238I4  0:4238I5  0:4238I6 2:2318:


Matrix form (8.58)


1 0
qn1
3:376
B qn C B
B 2 C B 1:0017
C
B
@ qn3 A @ 0:6309
0:4238
qn
0

0:7236
1:7363
0:6309
0:4238

1 0 I4 1
0:7236 0:2895
BI C
1:0017 0:198 C
5C
CB
B C:
4:9
0:379 A @ I6 A
0:4238 2:2318
1

8.4 Input-Output Conformity

273

We check non-uniform coordinates (8.59)


3:376  0:4921  0:7236  0:8654  0:7236  0:354  0:2895
0:4238  0:4921  0:4238  0:8654  0:4238  0:354 2:2318
0:4895

0:3254;
1:5065
1:0017  0:4921 1:7363  0:8654  1:0017  0:354  0:198
m12
1:5065
0:457

0:3022;
1:5065
0:6309  0:4921  0:6309  0:8654 4:9  0:354  0:379
m13
1:5065
0:5069

0:3367:
1:5065

m11

Load conductivities (8.60)


0:4146  0:3254
0:2;
0:3254  1
0:5775  0:3022
0:25;

0:3022  1
0:3284  0:3367
0:1666:

0:3367  1

1
YL1

1
YL2
1
YL3

References
1. Penin, A.: Invariant properties of cascaded six-pole networks. Int. J. Circuit Syst. Signal
Process. 6(5), 305314 (2012)
2. Penin, A., Sidorenko, A.: Method for transmitting signals through the three- wire direct current
line. MD Patent 692, 31 Oct (2013)
3. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer
Science + Business Media, Dordrecht (2014)

Chapter 9

Generalized Equivalent Circuit


of a Multi-port

9.1
9.1.1

Generalized Equivalent of an Active Two-Port


Disadvantages of Known Equivalent

Let us consider, for example, a familiar active two-port network in Fig. 9.1 with
changeable load conductivities YL1, YL2.
We rewrite the main relationships of Sect. 6.1.3. These active two-port is
described by the system of equations


I1
I2

    SC 
Y12
V1
I

1SC :
Y22
V2
I2

Y11
Y12

I1

y1N

1.25

y0N
V0
5

2.5

y2N

N
yN

9:1

0.833

0.625

V1

y1
0.25

y2
0.25

I2

YL1
V2
YL2

Fig. 9.1 Circuit with two load conductivities

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_9

275

276

9 Generalized Equivalent Circuit of a Multi-port

where Y parameters
Y11 y1 y1N 
Y12 y2N

y1N
;
yR

y21N
;
yR

yR y0N yN y1N y2N ;

Y22 y2 y2N 

y22N
:
yR

The short circuit SC currents of both loads


y1N
y2N
V0 ; I2SC Y20 V0 y0N
V0 :
I1SC Y10 V0 y0N
yR
yR

9:2

Expression (9.1) shows that the active two-port represents a passive part, which
SC
is set by conductivities of Y parameters, and two sources of currents ISC
1 , I2 , as it is
illustrated in Fig. 9.2 [1].
We may notice that SC currents are set by parameters Y10, Y20 which depend
practically on all elements of this circuit except y1, y2.
Therefore, at a possible change, for example, of conductivity yN, it is necessary
to make the recalculation of these SC currents what is inconvenient. The conductivity yN can be a part of a possible third load. In turn, the conductivity y1N belongs
more to the rst load, but also influences on the SC current of the second load. And
mutually, conductivity y2N influences on the SC current of the rst load.
All these features determine inconveniences for the evaluation of circuit characteristics, complicates regime calculations. In this sense, the parameters of the
generalized equivalent generator, proposed in Chap. 3, do not depend on a chosen
element of circuit.

9.1.2

Introduction of the Formal Variant of a Generalized


Equivalent

We use results of Sect. 3.3. At rst, let us consider the above two-port network
concerning the load YL1. Therefore, this circuit will be an active two-pole with
changeable element YL2 in Fig. 9.3.
The family of load straight lines I1(V1) is presented by expression (3.79)

I2

I1
V1

Y
SC

I1
YL1

V2
SC

I2

Fig. 9.2 Known equivalent circuit of an active two-port

YL2

9.1 Generalized Equivalent of an Active Two-Port

277

I1

V1

YL1

YL2

Fig. 9.3 An active two-pole with changeable element YL2

I1  I1G1 Yi1 V1 V1G1 :

9:3

The internal conductivity


Yi1 Y11 

Y12 2
:
YL2 Y22

9:4

The family of load straight lines shown in Fig. 9.4 represents a bunch of straight
lines with familiar center G1.
The bunch center coordinates correspond to expressions (3.77), (3.78), (3.80), or
(6.26), (6.27). Therefore, we get
V1G1 

Y20
y0N
V0 
V0 ;
Y12
y1N



y1
I1G1 y0N 1
V0 ;
y1N

9:5

I1
G1

YL1

YL1

G1

I1

G1

SC

I1

I1
OC

V1

V1
G1

V1

V1

YL2

YL2=

YL2= 0

YL2

Yi1
Fig. 9.4 Family of load straight lines with the parameter YL2

Yi1

278

9 Generalized Equivalent Circuit of a Multi-port


G1

VL

I1
V1

G1

IL

Yi

YL1

Fig. 9.5 Generalized equivalent generator to the load YL1

G1
YL1

I1G1
y1 y1N :
V1G1

9:6

Let us convert expression (9.3) to the relative form




I1
Yi1 V1
 1  G1
1 :
I1G1
YL1 V1G1

9:7

G1
G1
Thus, the values IG1
are scales for corresponding values.
1 , YL1 , and V1
Expressions (9.3) and (9.7) determine a generalized equivalent generator in Fig. 9.5.
We remind that the value Y0i1 = YG1
L1 is the characteristic value for the conductivity Yi. In this case, the conductivity Yi voltage is equal and opposite to the
source VG1
1 voltage. Therefore, I1 = 0, V1 = 0 for all conductivity YL1 values.
Second, let us consider now the above two-port network concerning the load
YL2. This circuit will be an active two-pole with changeable element YL1 in Fig. 9.6
too.
Similar relationships are obtained for the second load YL2

I2  I2G2 Yi2 V2 V2G2 ;


V2G2 

y0N
V0 ;
y2N

Yi2 Y22 



y2
I2G2 y0N 1
V0 ;
y2N

G2
YL2
y2 y2N :

I2
V2

A
YL1

YL2

Fig. 9.6 An active two-pole with changeable element YL1

Y12 2
:
YL1 Y11

9:8
9:9

9.1 Generalized Equivalent of an Active Two-Port

G1

V1

279

G2

V2

G2

G1

I2

I1
Yi1

YL1

Yi2

YL2

Fig. 9.7 Formal equivalent generator as the association of the two active two-poles

These expressions determine also the similar generalized equivalent generator.


Third, the common or formal scheme of the equivalent generator of our active
two-port network may be constituted by Fig. 9.7 [2].
This scheme gives an evident representation about mutual influence of loads and
allows carrying out regime calculations.

9.1.3

Introduction of the Principal Variant of a Generalized


Equivalent Circuit

It is possible also to obtain one more scheme of the equivalent generator. This
principal variant is demonstrated by Fig. 9.8 [2].
To prove that, we consider expression (6.43) for the rst input conductivity
Y11

I1G1  I1SC
J1

:
U1
V1G1

This expression denes the input conductivity for the passive part of two-port
network at the short circuit for the rst load and short circuit for the second pair of
terminals in Fig. 9.9a.
The similar relationship is for the second input conductivity
Y22

I2G2  I2SC
J2

:
U2
V2G2

G1

V1 /2

G2

U1

V2 /2

U2

V1

V2

Y
I1
YL1

J1
G1

I1 /2

I2

J2
G2

I2 /2

YL2

Fig. 9.8 Electrical scheme of the generalized equivalent generator of the active two-port

280

9 Generalized Equivalent Circuit of a Multi-port

Fig. 9.9 a Denition of the


rst input conductivity of the
passive part of two-port
network, b denition of the
second input conductivity

(a) V G1

U1

U2=0
Y

SC

I1

G1

J1

I1

(b)
U1=0

Y11

G2

V2

U2
Y
J2
Y22

G2

I2

SC

I2

Analogously, we have the input conductivity for the passive part of two-port
network at the short circuit for the second load and short circuit for the rst pair of
terminals in Fig. 9.9b.
The association of these two schemes leads to the above scheme in Fig. 9.8.
Taking into account the superposition theorem, the values of all constituent sources
of current and voltage are decreased twice. The calculation proves such a scheme
for equivalent generator.
Next, we may obtain the system of equations which describes the proposed
equivalent generator. Using the specied designations in Fig. 9.8, it is possible to
write down
V1G1
V1 ;
2
G2
V
U2 2 V2 ;
2

U1

I1G1
 I1 ;
2
G2
I
J2 2  I 2 :
2
J1

9:10

The following relationships take place for the passive part of two-port network
J1 Y11 U1  Y12 U2
J2 Y12 U1 Y22 U2 :

9:11

Then, taking into account (9.10), the expression is obtained




V1G1
V2G2 I1G1
I1 Y11 V1 Y12 V2 Y11
Y12

:
2
2
2

9:12

9.1 Generalized Equivalent of an Active Two-Port

281

Let us compare this form to expression (9.1). Then, the term in the brackets is
equal to the SC current ISC
1 . In addition, the structure of this short circuit current for
(9.12) demonstrates its components.
The component Y11VG1
1 /2 corresponds to the own current of the two-port network and depends on its parameters.
The component IG1
1 /2 is dened by the constituent current source.
The component Y12VG2
2 /2 corresponds to a mutual current on account of the
constituent voltage source for the second pair of terminals.
The expression for the current I2 is obtained similarly to (9.12)


V2G2
V1G1 I2G2
Y12

I2 Y22 V2 Y12 V1 Y22


:
2
2
2

9:13

The offered scheme of the equivalent generator is convenient at the modeling of


an active two-port network.
The values of the constituent voltage and current sources of the generalized
equivalent generator do not depend on the conductivity yN in Fig. 9.1. This feature
represents a practical interest. This conductivity may be a part of the third
changeable load. Therefore, Y parameters of the passive part of two-port are only
recalculated.
Besides, the bunch center parameters of one load do not depend on the conductivities entering into the circuit of other load. For example, the center G1 is
dened by the elements y1, y1N ,and does not depend on y2, y2N.
Therefore, such property of this equivalent generator simplies analysis of a
circuit.
Example 1 We use the date of Example 2, Sect. 6.1.3. For the running regime,
expressions (9.10) have the values
10
V1 6:958;
2
15
V2 9:149;
U2
2
U1

15
 I1 6:521;
2
16:25
 I2 7:3:
J2
2
J1

We check relationships (9.11)


J1 Y11 U1  Y12 U2 1:2  6:958  0:2  9:149 6:52;
J2 Y12 U1 Y22 U2 0:2  6:958 0:95  9:149 7:3:
SC currents (9.12), (9.13)
I1SC 1:2  5 0:2  7:5 7:5 3;
I2SC 0:95  7:5 0:2  5 8:125 2:

282

9.2

9 Generalized Equivalent Circuit of a Multi-port

Generalized Equivalent of an Active Three-Port

Let us use the above approach for an active three-port network with changeable
load conductivities in Fig. 9.10. We use now the results of Sect. 6.2. This network
is described by the following system of equations
2

3 2
Y11
I1
4 I2 5 4 Y12
I3
Y13

Y12
Y22
Y23

3 2 3 2 SC 3
V1
Y13
I1
Y23 5  4 V2 5 4 I2SC 5:
Y33
V3
I3SC

9:14

Taking into account the voltages V1 = I1/YL1, V2 = I2/YL2, and V3 = I2/YL3, we get
three bunches of planes. The intersection of planes of one bunch among themselves
denes a bunch axis.
The point of intersection of this bunch axis with the axis I1 is dened by (6.47)


y1
I1G1 V0 y0N 1
;
y1N

V1G1 

y0N
V0 ;
y1N

G1
YL1
y1N y1 :

9:15

This point gives the center G1. Similarly, we get the center G2 by (6.45)


y2
I2G2 V0 y0N 1
;
y2N

Fig. 9.10 Active multi-port


with three loads

V2G2 

y0N
V0 ;
y2N

G2
YL2
y2N y2 :

9:16

I1

y1N

V1
y1

y0N

I2

y2N

YL1
V2

V0

y2
YL2

I3

y3N

V3
y3
YL3

9.2 Generalized Equivalent of an Active Three-Port

283

Also, the center G3 by (6.46)




y3
I3G3 V0 y0N 1
;
y3N

V3G3 

y0N
V0 ;
y3N

G3
YL3
y3N y3 :

9:17

Using these parameters of the centers, we get form (6.55) of system (9.14)


I1
V1
I1SC
I SC V2
I SC V3
I SC

1

 1G1 G2  1G1 G3 1G1 ;
G1
G1
G1
I1
V1
I1
I1 V 2
I1 V3
I1


SC
SC
SC
I2
I V1
V2
I
I V3
I SC
 2G2 G1 G2 1  2G2  2G2 G3 2G2 ;
G2
I2
I2 V 1
V2
I2
I2 V3
I2


SC
SC
SC
I3
I3 V1
I3 V2
V3
I3
I3SC




1

:

I3G3
I3G3 V1G1 I3G3 V2G2 V3G3
I3G3
I3G3

9:18

From here, we may express Y parameters


Y11
Y22

I1G1  I1SC
;
V1G1

I G2  I SC
2 G22 ;
V2

Y12
Y23

I1SC
I2SC

;
V2G2 V1G1

I SC
I SC
2 G3 3 G2 ;
V3
V2

Y13
Y33

I1SC
I3SC

;
V3G3 V1G1

I G3  I SC
3 G33 :
V3

9:19

Let us obtain the generalized equivalent generator similar to Fig. 9.8. For this
purpose, we consider also expression (9.19) for the rst conductivity
Y11

I1G1  I1SC
J1

:
G1
U1
V1

This expression denes the input conductivity for the passive part of three-port
network at the short circuit for the rst load and short circuit for the second and
third pairs of terminals in Fig. 9.11a.
The similar formulas for the second and third conductivities have the view
Y22

I2G2  I2SC
J2

;
G2
U2
V2

Y33

I3G3  I3SC
J3

:
G3
U3
V3

These relationships correspond to Fig. 9.11b, c.


The association of these three schemes leads to the scheme in Fig. 9.12. Taking
into account the superposition theorem, the values of all constituent sources of
current and voltage are decreased thrice [3]. The calculation proves such a scheme.

284

9 Generalized Equivalent Circuit of a Multi-port

Fig. 9.11 a Denition of the


rst input conductivity of the
passive part of three-port
network, b denition of the
second input conductivity,
c denition of the third input
conductivity

(a) V G1

U1

SC

I1

U2=0

Y11

U3=0

J1

G1

I1

(b)
U1=0

G2

U2

V2

Y
J2

U3=0

I2

I2

(c)
U1=0

SC

G2

Y22

G3

U3

V3

Y
J3

U2=0

I3

I3

Fig. 9.12 Generalized


equivalent generator of the
active three-port

G1

V 1 /3

U1
Y

SC

G3

Y33

J1

I1
G1

I 1 /3

V1
YL1

G2

V 2 /3

U2
J2

I2
G2

I2 /3

V2
YL2

G3

U3

V 3 /3
V3

J3

I3
G3

I 3 /3

YL3

9.2 Generalized Equivalent of an Active Three-Port

285

We obtain the system of equations which describes the proposed equivalent


generator. Using the specied designations in Fig. 9.12, it is possible to write down
V1G1
V1 ;
3
V G2
U2 2 V2 ;
3
V3G3
V3 ;
U3
3

U1

I1G1
 I1 ;
3
I G2
J2 2  I 2 ;
3
I3G3
 I3 :
J3
3
J1

9:20

The following relationships take place for the passive part of three-port network
J1 Y11 U1  Y12 U2  Y13 U3 ;
J2 Y12 U1 Y22 U2  Y23 U3 ;

9:21

J3 Y13 U1  Y23 U2 Y33 U3 :


Then, taking into account (9.20), the current is obtained
I1 Y11 V1 Y12 V2 Y13 V3


V G1
V G2
V G3 I G1
Y11 1 Y12 2 Y13 3 1 :
3
3
3
3

9:22

The term in the brackets is equal to the SC current ISC


1 . This term shows the
components of this current.
The expressions for the currents I2, I3 are obtained similarly
I2 Y12 V1  Y22 V2 Y23 V3


V1G1
V2G2
V3G3 I2G2
 Y22
Y23

Y12
;
3
3
3
3

9:23

I3 Y13 V1 Y23 V2  Y33 V3




V G1
V G2
V G3 I G3
Y13 1 Y23 2  Y33 3 3 :
3
3
3
3

9:24

The values of the constituent voltage and current sources of the generalized
equivalent generator do not depend on the conductivity yN in Fig. 9.10. It represents
a practical interest. This conductivity can be a part of the fourth changeable load.
Besides, the bunch center parameters of one load do not depend on the conductivities entering into the circuit of other loads. For example, the center G1 is
dened by elements y1, y1N and does not depend from y2, y2N, and y3, y3N.

286

9 Generalized Equivalent Circuit of a Multi-port

Example 2 We use the date of Example 3, Sect. 6.2. Expressions (9.20) give the
values
10
15
1:948 5:28133; J1
 0:974 4:026;
3
3
15
16:25
U2
1:64 6:64; J2
 0:82 4:596;
3
3
9:377
14:844
U3
1:61 4:7356; J3
 1:61 3:338:
3
3

U1

We check relationships (9.21)


J1 1:236  5:2813  0:176  6:64  0:282  4:7356 4:026;
J2 0:176  5:2813 0:966  6:64  0:188  4:7356 4:596;
J3 0:282  5:2813  0:188  6:64 1:283  4:7356 3:338:
SC currents (9.22)(9.24)


I1SC
I2SC
I3SC


10
15
9:377 15
0:176
0:282

1:236
2:641;
3
3
3
3


10
15
9:377 16:25
0:188

0:176  0:966
1:761;
3
3
3
3


10
15
9:377 14:844
0:188  1:283

0:282
2:817:
3
3
3
3

References
1. Bessonov, L.A.: Teoreticheskie Osnovy Elektrotekhniki. Elektricheskie tsepi, Izd.9. (Basic
Electrical Engineering Theory: Electric Circuits, 9th edn.). Vyshaia shkola, Moskva (1996)
2. Penin, A.: About the denition of parameters and regimes of active two-port networks with
variable loads on the basis of projective geometry. WSEAS Trans. Circuits Syst. 10(5), 157
172 (2011). http://www.worldses.org/journals/circuits/circuits-2011.htm. Accessed 30 Nov
2014
3. Penin, A.: Parameters and characteristics of the modied equivalent generator of an active
multiport network. Elektrichestvo 5, 3239 (2012)

Part III

Circuits with Non-Linear


Regulation Curves

Chapter 10

Regulation of Load Voltages

10.1

Base Model. Display of Conformal Geometry

In power supply systems with limited voltage source power, the restriction of load
power, two-valued regulation characteristic, and interference of several loads is
observed [15]. Distributed power supply systems, autonomous or hybrid power
supply systems with solar cells, fuel elements, and accumulators may be examples
of such systems [8].
At the present time, a digital control of voltage converters is used. One way of
the digital control performance is a predictive technique. In one switching period, a
duty cycle for the next switching cycle is calculated, based on a sensed or observed
state and input/output information [2, 3].
Also, a feed-forward control method improves a load regulation dynamics of
converter [1]. This method calculates the required duty ratio variation by a predicted load current.
Therefore, it is necessary to take into account an internal resistance of power
supply, to carry out analysis of the load interference and obtain relationships for
denition of regime parameters at the possible coordinated predictive control for
preset load regimes. To simplify the solution of task and to reveal basic moments of
this influence, it is expedient to consider static regulation characteristics and idealized models of voltage converters.
Anyway, we consider a power supply system with two regulated voltage converters (or voltage regulators VR1, VR2) and given loads R1, R2 in Fig. 10.1 [911,
13]. Generally, voltage converters with switched tapped secondary windings of
transformers, multicell or multilevel voltage converters, pulse-width modulation
PWM converters, and so on may be examples of these regulators.
The regulators dene transformation ratios n1, n2
n1

V1
;
V

n2

V2
:
V

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_10

10:1

289

290

10

Fig. 10.1 Power supply


system with regulation of load
voltagesof load voltages

Regulation of Load Voltages

VR1
Ri
V0

V1
R

I0

VR2

n1
V2
R

n2
The interference of the regulators on the load voltages V1, V2 is observed
because of an internal resistance Ri.
Let us obtain equations describing behavior of this circuit at change of the values
n1, n 2.
From (10.1) it is evident
V1 n1 V;

V2 n2 V:

The total load power


PL

1
1
V1 2
V2 2
R1
R2


1
1
n1 2
n2 2 V2 :
R1
R2

On the other hand,


PL I0 V

V0  V
V:
Ri

Finally, we get the required equation




Ri
Ri
V0 2 V0 2
:
V1 2
V2 2 V 

R1
R2
2
4

10:2

This expression represents a sphere by the coordinates V1, V2, V in Fig. 10.2. For
simplication of drawing, the axes V1, V2 are superposed.
In turn, the variables n1, n2 are resulted by a stereographic projection of spheres
points from the pole 0, 0, 0 [14]. These variables dene the conformal plane n1n2
[4] the axes n1, n2 are superposed too.

10.1

Base Model. Display of Conformal Geometry

Fig. 10.2 Stereographic


projection of spheres points
on a tangent plane n1n2

291

V
V0

n1

n1M

n2

n2M

n1
n2

n1
n2

V0
2

V1 V1M

V1

V2 V2M

V2

In the plane V1, V2 we have an area of voltage changes. This area is dened by
the internal area of circle (ellipse) in Fig. 10.3a and corresponds to the spheres
equator. In this case V = V0 / 2. Taking into account (10.2), we obtain the equation
of this circle
Ri
Ri
V0 2
:
V1 2
V2 2
R1
R2
4

10:3

In the plane n1n2 this equation has the form


Ri
Ri
n1 2
n2 2 1:
R1
R2

10:4

Then, we get the maximum load voltages


V1M

V0

2

r
R1
;
Ri

V2M

V0

2

r
R2
:
Ri

10:5

Similarly, the maximum transformation ratios


n1M 

r
R1
;
Ri

n2M 

r
R2
:
Ri

10:6

Let, for example, the rst load voltage regime V1 = const (there is a line L1 in
Fig. 10.3a) be supported by n1, n2 changes. Then, a spheres circular section L1 and
corresponding circle L1 in the plane n1n2 (there is a line L1 in Fig. 10.3b) turns out.

292

10

Fig. 10.3 a Lines L1, K1 of


characteristic regimes in the
plane V1V2. b The same lines
in the conformal plane n1n2

Regulation of Load Voltages

V2

(a)
V2M
2

k2
1

k2

K1
1
-V1

-V1M

L1
1

V1M V
1

V1 V1

n2

(b)

n2 M
2

k2

V1

k2

V1

n1 M
2

1
n1 n1

n1
L1

K1

Let us obtain the equation of this circle L1.


From (10.1) it is evident
V1 2 n1 2 V 2 ;

V2 2 n2 2 V 2 :

Then, by (10.2)
V

V0
1

Ri
R1

Ri
R1

n1 2

Ri
R2

n2 2

Ri
R2

n2 2

10:7

10:8

Therefore, we have
V1

n1 V 0
n1 2

10.1

Base Model. Display of Conformal Geometry

293

Thus, the equation of the circle L1 has the view


Ri
Ri
V0
n2 2
n1 2  n1
1 0;
R2
R1
V1

10:9

where the voltage V1 is a parameter.


The points of intersection of the circle L1 with the axis n1 correspond to
Eq. (10.9) as n2 = 0. Then
n1 2  n1

R1 V0 R1

0:
Ri V1
Ri

Roots n1 ; ~
n1 of this equation are characterized by the inverse property
n1  ~n1

R1
n1M 2 :
Ri

Therefore, in the plane n1n2, the two families of circles L1 correspond to the
parallel straight lines in the plane V1V2 for different values V1. These two families of
circles have the centers n1M. In turn, the values n1M conform to maximum
voltages V1M.
Such families of circles L1 describe the regime V1 = const as the rotation group
of sphere, as it is shown by arrows in Figs. 10.2 and 10.4.
This motion of points n1 n2 has the two xed points n1M and it is the elliptic
projectivity in the form
n2

an1 b
;
cn1 d

10:10

where coefcients a, b, c, d are parameters [5, 6]. The initial point n1 = n11 + jn12 and
subsequent point n2 = n21 + jn22 are complex values.
Fig. 10.4 Regime V1 = const
as the rotation group of sphere

n2

n1
L1
K1

294

10

Fig. 10.5 a Rotation of the


equatorial plane of sphere in
the coordinate plane V, V2
gives the circle family K1.
b Moving of points of the
circles K1 as a hyperbolic
transformation

Regulation of Load Voltages


V

(a)
V0

V1=0
1

k2

k2
K1
V2

(b)

n2

n1

K1

Let a set of initial points be situated on the spheres equator for different values
of V1. Projective transformation (10.10) shifts each initial point along its circle L1.
In turn, all subsequent points constitute an arc of circle K1 in Fig. 10.4. This circle
K1 is orthogonal to all circles of family L1.
Let us consider this shift in detail. The rotation of the equatorial plane of sphere
about its diameter, shown by arrows in Fig. 10.2, gives the circle family K1 with
parameters k12, k22, and so on in Fig. 10.3. This rotation by the coordinates V V2 is
shown in Fig. 10.5a.
In turn, in the plane V1V2, the projected circles k12, k22 give the family of ellipses
K1, as shown in Fig. 10.3a. Let us obtain the equation of this ellipse.
By Fig. 10.5a, we have


V0
V 2 k2 V 
n2 V;
2
where k2 is an angular coefcient. Therefore

V

V0
2

V2
:
k2

10:11

10.1

Base Model. Display of Conformal Geometry

295

Substituting this expression in (10.2), we get


!
Ri
Ri
1
V0 2
2
2
:
V1

V
2
R1
R2 k2 2
4

10:12

Next, we obtain the equation of corresponding circles in the plane n1n2. From
(10.11) it follows that
V0
k 2  n2
2
:
V
k2
Using (10.7), we get the required equation of circle K1
Ri
Ri
n2
n2 2
n1 2 2  1 0:
R2
R1
k2

10:13

This expression corresponds to (10.4) as k2 = . The moving of points of the


circles K1 is shown by arrows in Fig. 10.5b. Such moving, as a hyperbolic transformation with the two xed points n1M, has form (10.10) too.

10.2

Using of Hyperbolic Geometry Model

At the increase of the values n1, n2 on some step of switching cycle, a running point
may pass over the equator and the voltage V2 is going down that is inadmissible.
Therefore, it is better to use such groups of transformations or movements of points
in the plane n1n2, when it is impossible to move out the running point over the circle
or equator of sphere. So, we must decrease the next values n1, n2 by some rule.
In this sense, we come to hyperbolic geometry [7]. There is Poincares model in
the plane n1, n2 shown in Fig. 10.6. The corresponding circle carries the name of
absolute and denes an innitely remote border. The equation of absolute conforms
to (10.4).
Fig. 10.6 Poincares model
of hyperbolic geometry

n2
L1

K1

n1

296

10

Regulation of Load Voltages

The arcs of circles L1 are straight lines of this model. In turn, the arcs of circles
K1 are the lines too. The distance between these lines is constant. The straight lines
L1 are orthogonal to these equidistant lines K1.
In turn, the Beltrami Klein model is the other hyperbolic geometry model in the
plane V1V2 in Fig. 10.3a. The lines L1, K1 of this model have the same sense.
Now, we may obtain required expressions or rules of regime parameters and
their changes.

10.2.1 Case of One Load


At rst, we consider our power supply system with one load, as n2 = 0, V2 = 0.
Using (10.8), we get the regulation characteristic equation
V1

n1 V 0
1

Ri
R1

n1 2

For example, this regulation characteristic is shown in Fig. 10.7a as Ri = R1.


Let us consider transformation (10.10). In the given case, this expression is a
projective transformation for the real variable n1. The regime change goes only on
the axes V1, n1 as it is illustrated by arrows. Points n11, n21 of initial and subsequent
regime form a segment n11n21. The moving of this segment for different initial points
and the conformity of variables n1, V1 is shown in Fig. 10.7b.
It is obvious that the Euclidean length of segment n11n21 is being decreased, while
this segment approaches to the xed points n1M of the absolute.
We know that a projective transformation has an invariant in the form of
cross-ratio for four points. In the given case, there are these two xed points (as the
base points), point n11 of initial regime, and point n21 of subsequent one.
Then, the cross-ratio m21
n , which corresponds to the regime change, has the form [12]
2 1
m21
n n1M n1 n1 n1M

n21 n1M n11 n1M



:
n1M  n21 n1M  n11

10:14

The value of this cross-ratio is constant for different initial points, as it is shown
in Fig. 10.7b.
It is also possible to constitute the cross-ratio m1n for the initial or running regime
1
n1 relatively to the origin of coordinate n1 = 0; that is,
m1n n1M n11 0 n1M

n11 n1M
:
n1M  n11

10:15

10.2

Using of Hyperbolic Geometry Model

Fig. 10.7 a Regulation


characteristic. b Moving of a
segment for different initial
points

297

(a) 0.5

V1
V0

0.0

n1M
-0.5
-2

-1

n1
21

21

(b)
-V1M

V1
21

21

21

n1

mn

n1

mn

mn

mn
1

n1M
21

21

mn

V1

n1
2

n1

V1M
21

n1
0

V1
2

V1

n1
-n1M

21

V1

V1

mn

The conformity of the values n11, m1n is shown in Fig. 10.7b too.
The value m1n determines the non-uniform projective coordinate of the value n11.
Hence, the point n1 = 0 is a unite point.
In the same way, the cross-ratio for the subsequent regime n21
m2n n1M n21 0 n1M

n21 n1M
:
n1M  n21

10:16

Obviously, a group property is executed; a subsequent regime value is equals to


initial regime value multiplied by regime change value.
Let us obtain transformation (10.10) for our case. To do this, using normalized
values
n21

n21
;
n1M

n11

n11
;
n1M

298

10

Regulation of Load Voltages

we get (10.14) in the form


21 n11 1
m21
n 1 n

n21 1 n11 1

:
1
n21 1  n11

10:17

Then, the subsequent value


n21

21
n11 m21
n  1=mn 1
:
21
1 n11 m21
n  1=mn 1

10:18

m21
n 1
:
m21
n 1

10:19

n11 n21
1
:
1 n11 n21
1

10:20

We may introduce a value


n21
1
Finally, we obtain
n21

There is a required property of this transformation. If the initial value n11 1,


then the subsequent value n21 1 for various values n21
1 .
In turn,
n21
1

n21  n11
:
1  n21 n11

10:21

So, there is a strong reason to introduce the transformation ratio change as n21
1 .
Let us now consider the variable V1. Then, the cross-ratio mV for the initial
regime V11 and subsequent regime V21 has the form
m1V V1M V11 0 V1M

V11 V1M
;
V1M  V11

m2V

V12 V1M
:
V1M  V12

10:22

The following equality takes place:


mV mn 2 :

10:23

This expression leads to identical values if we use the hyperbolic metric to


determine a regime value
S Ln mV 2Ln mn :

10:24

10.2

Using of Hyperbolic Geometry Model

299

Using normalized values


2
12 V1 ;
V
V1M

1
11 V1 ;
V
V1M

we get (10.8) in the form


1
V

2n1
1 n1 2

10:25

Similarly to (10.20) and (10.21), it follows that


21
1
12 V1 V1 ;
V
1

1 V1 V121

10:26

12  V
11
V
2
11 :
1 V
1V

10:27

V121

Thus, there is a strong reason to introduce the value of voltage change as V21
1 .
21
The validity of such denition for the changes n21
1 and V1 is conrmed by
expression (10.25); that is,
V121

2n21
1
2
1 n21
1

10:28

Therefore, a concrete kind of circuit and character of regime determines system


parameters.
Hence, arbitrary expressions are excluded.

10.2.2 Case of Two Loads


Moving of points in the plane n1 n2
Let us consider the kinds of points move in the plane n1 n2 with the aid of Fig. 10.6.
The arcs of lines L1, K1 are virtual trajectories of regime changes. In turn, points of
initial and subsequent regimes are labeled on these lines, as it is shown in Fig. 10.8.
The family of lines L1 corresponds to the voltages or parameters V11, V21, and so
on. These lines intersect the lines K1 with parameters k12, k22, and so on.
As the result, we have the sets of points on the lines L1. For example, there are
points C1n, D1n for the line with V11 parameter.

300

10

Fig. 10.8 Lines L1, K1 as


virtual trajectories of regime
change

Regulation of Load Voltages

n2

1n

2n

k2

1n

2n

2
1

k2

3n

2n

1n

2n

3n

2n

F3n

1n

K1

1n

2n

3n

0
2

V1

V1

n1

A
L1

3n

2n

A1n

System of Eqs. (10.9) and (10.13)


8
Ri
Ri
V0
2
>
>
n1 2  n1
1 0
< n2
R2
R1
V1
R
R
n
>
>
: i n2 2 i n1 2 2 2  1 0 :
R2
R1
k2
determines coordinates of these points.
This system gives the quadratic equation for n1
"

V0
k2
2V1

2

#


Ri
Ri
V0
2
2 Ri

k2
1 n1
n1 
R2 R1
V1
R2


2 Ri
k2
1 0:
R2

In turn,


n2 k 2


V0
1  n1
:
2V1

10:29

10:30

Further, we must determine the coordinates of points A1, F1 with the aid of
Eqs. (10.4) and (10.9)

Ri
R2
Ri
R2

n2 2
n2 2

Ri
R1
Ri
R1

n1 2  n1 VV01 1 0
n1 2  1 0:

10.2

Using of Hyperbolic Geometry Model

301

Then
V1
n1 2 ;
V0

s


R2 R2
V1 2
:
n2 

2
Ri R1
V0

10:31

Moving of points along the line L1


Let us constitute the cross-ratio m21
nL1 that corresponds to the regime change; that is,
points C1n, D1n of line L1 and line K1 with parameters k12, k22.
Then
m21
nL1 A1n D1n C1n F1n

D1n  A1n C1n  A1n



:
D1n  F1n C1n  F1n

10:32

The coordinates of the points A1n, C1n, D1n, F1n are dened by normalized
complex values
nA1 nA1
nA1
1 j
2 ;

nC1 nC1
nC1
1 j
2 ;

nD1 nD1
nD1
1 j
2 ;

nF1 
nF1
nF1
1 j
2 :

In particular, the coordinates of A1, F1, are conjugate complex values


nA1
nF1
1 
1 ;

jnA1
nF1
2 j
2 ;

2
2
 1 nF1
where 
nF1
nF1
1 V1 ; 
1 
2 1:
Then, the cross-ratio has the form

nA1 nD1 nC1 nF1


m21
nL1 

nD1  
nA1 nC1  nA1

:
nD1  
nF1 nC1  nF1

10:33

The result of calculation shows that this cross-ratio is a real value. The
cross-ratio m21
nL1 corresponds to the points C2n, D2n and so on.
Let us determine transformation (10.10) for the line L1 with parameter V11. Using
(10.33), we get the subsequent value

nD1

 F1
 C1
21
n1 jnF1
1 m21
n 1
nL1 =mnL1  1 
 2F1
;
21 =m21  1
nC1  n1  jnF1
1

m
2
nL1
nL1

10:34

where 
nC1 is the initial value.
In particular, when the initial and subsequent value situated on the axis n2 , this
expression has the form

302

10

nD1
2

Regulation of Load Voltages

21
21
nC1
2 1 mnL1 =mnL1  1
:
21
21
1 nC1
2 1 mnL1 =mnL1  1

This transformation conforms to (10.18) of one load.


Moving of points along the line K1
Similarly, we have the sets of points on the lines K1. For example, there are points
C1n, C2n for the line with k12 parameter.
Let us constitute the cross-ratio m21
nK1 that corresponds to the regime change; that
is, points C1n, C2n
C2n 1 C1n 1

:
C2n  1 C1n  1

m21
nK1 1 C2n C1n 1

10:35

The coordinate of C2n is dened by normalized complex value


nC2 nC2
nC2
1 j
2 :
Then, the cross-ratio has the form
C2 nC1 1
m21
nK1 1 n

nC2 1 nC1 1

:
1
nC2 1  nC1

10:36

The result of calculation shows that this cross-ratio is a real value. The
cross-ratio m21
nK1 corresponds to points, D1n, D2n and so on.
Let us determine transformation (10.10) for line K1. Using (10.18) and (10.19),
we get the subsequent value
nC2

nC1 n21
K1
;
1 nC1 n21
K1

10:37

m21
nK1  1
m21
nK1 1

10:38

where the real value


n21
K1

determinates the transformation ratio change.


2
1 V
Moving of points in the plane V
Points of initial and subsequent regimes are labeled on the lines L1, K1 as it is
shown in Fig. 10.9.

10.2

Using of Hyperbolic Geometry Model

303

Fig. 10.9 Lines L1, K1 as


virtual trajectories of regime
changes

V2
k

F1V

F2V

2
2

k2

3V

k2

1V

1V

1V

K1

1V

2V

2V

E
D

3V

3V

2V

2V

3V

3V

V1
V

1
1

V1

A1V

A2V

A3V

Moving of points along the line L1


Similarly to (10.32), let us constitute the cross-ratio
m21
VL1 A1V D1V C1V F1V

D1V  A1V C1V  A1V



:
D1V  F1V C1V  F1V

10:39

The coordinates of the points A1V, D1V, C1V, F1V are dened by normalized
complex values
 A1 V
11 jV
2A1 ;
V
11 jV
2C1 ;
 C1 V
V

 D1 V
11 jV
2D1 ;
V
 F1 V
11 jV
2F1 :
V

10:40

In particular, the coordinates of A1V, F1V are conjugate complex values


2F1 :
2A1 jV
V
Then, cross-ratio (10.39) has the form

m21
VL1

 D1
V

 C1
V


V
2


V
2

 D1 V
 C1 V
 F1 V
 F1 V2F1 1 V2F1 1
V
:
 V2C1
 2D1 2F1   2C1 2F1 V2D1
V2  V2
V2  V2
2
2
F1  1
F1  1

10:41

304

10

Regulation of Load Voltages

The result of calculation shows that equality (10.23) takes place


21 2
m21
VL1 mnL1 :

10:42

Let us determine transformation (10.10) for line L1 with parameter V11. Using
(10.41), (10.36) and (10.37), we get the subsequent value
 C1
V

21
2
2D1
 F1 VL1
V
V
2

;
 C1
V
 F1
21
V
2
1 V2F1 VL1

10:43

where the value


 C1
V
2

 D1
V
2

21
VL1

 F1  V F1
m21  1
V
VL1
2 V D1 V2C1
21
mVL1 1 1  2F1 2F1

10:44

 V

V
2
2

2 .
determinates the voltage change for V
Moving of points along the line K1
Similarly to (10.36) let us constitute the cross-ratio
2 1
m21
VK1 1 V1 V1 1:

10:45

The result of calculation shows that equality (10.23) and (10.42) takes place
21 2
m21
VK1 mnK1 :

1 .
Let us determine the transformation which similar to (10.26) for voltage V
Then
21
1
12 V1 VK1 ;
V
21
11 VK1
1V

10:46

where the value


21
VK1

12  V
11
V
m21
VK1  1

11
12 V
m21
1V
VK1 1

1 .
determinates the voltage change for V

10:47

10.2

Using of Hyperbolic Geometry Model

305

2 . To do this, we use
Now, let us determine the transformation for voltage V
ellipse Eq. (10.12) for normalized values
!
R2 1
V1 1
V2 2 1:
Ri k2 2
2

10:48

11 ; V
21 , V
12 ; V
22 of the initial C1V and subsequent
Substituting the coordinates V
C2V points into (10.48), we get the system of equation
8
!
>
R2 1
>
1 2
>

21 2 1
>
V 1
V
>
< 1
Ri k2 2
!
>
>
R
1
2
2
>
2
2 2

>
>
: V1 1 Ri k 2 V2 1:
2
22
Using (10.46), we obtain the equation for V


21
11 VK1
V
21
11 VK1
1V

2

1 2
22 2 1  V1 1:
V
21 2
V

This equation gives the required subsequent voltage


22
V

q
21
V
21 2 :
1  VK1
21
11 VK1
1V

10:49

Obtained expressions (10.46) and (10.49), as transformation groups, dene a


1 ; that is, C1V C2V C3V and so on, step
parallel shift or sliding along the axis V
21
by step. The parameter VK1 is the value of this shift. In this case, we have a priority
1 . The parallel shift is one of the known three types of
control of the load voltage V
movement in the hyperbolic plane.
The obtained expressions may be generalized for three or more loads.

10.3

Example

Consider the circuit in Fig. 10.1. Let the parameters be given as follows:
V0 5;

Ri 1;

R1 1:25;

R2 2:

306

10

Regulation of Load Voltages

Expression (10.2) of sphere


0:8V1 2 0:5V2 2 V  2:52 6:25:
Equations (10.3) and (10.4) of absolute
0:8V1 2 0:5V2 2 6:25;
0:8n1 2 0:5n2 2 1:
Maximum load voltages (10.5)
p
V1M 2:5 1:25 2:795;

p
V2M 2:5 2 3:5353:

Maximum transformation ratios (10.6)


p
n1M  1:25 1:118;

p
n2M  2 1:4142:

Equation (10.9) of circle L1


0:5n2 2 0:8n1 2  n1

5
1 0:
V1

Equations (10.12) and (10.13) of ellipse and circle K1


!
2

V2 2 6:25;
k2 2
n2
0:5n2 2 0:8n1 2 2  1 0:
k2

0:8V1 0:5

10.3.1 Case of One Load


Let the initial and subsequent load voltage be equal to
V11 1;

V12 2:5:

Then, the corresponding transformation ratios


n11 0:2068;

n21 0:6909:

10.3

Example

307

Cross-ratio (10.14) of regime change


0:6909 1:118 0:2068 1:118

2:913:
0:6909  1:118 0:2068  1:118

m21
n

Cross-ratio (10.15) for the initial regime by n11


m1n

0:2068 1:118
1:4539:
1:118  0:2068

Cross-ratio (10.16) for subsequent regime by n21


m2n

0:6909 1:118
4:2353 2:913  1:4539:
1:118  0:6909

The normalized transformation ratios



n21

0:6909
0:618;
1:118

n11

0:2068
0:185:
1:118

Transformation ratio change (10.19)


n21
1

2:913  1
0:4888:
2:913 1

Projective transformation (10.20) for subsequent value


n21

0:185 0:4888
0:618:
1 0:185  0:4888

The next subsequent value n31 relatively to n21



n31

n21 0:4888
0:618 0:4888
0:85:

1 n21  0:4888 1 0:618  0:4888

and so on.
We see that actual values of the transformation ratio changes are decreasing at
each time.
Cross-ratio (10.22) for initial and subsequent regimes by voltages V11, V21
m1V

1 2:795
2:1142;
2:795  1

m2V

2:5 2:795
17:949:
2:795  2:5

308

10

Regulation of Load Voltages

Equality (10.23)
m1V 2:1142 1:45392 ;

m2V 17:949 4:23532 :

Hyperbolic metric (10.24)


S1 Ln 2:1142 2Ln 1:4539 0:7486;
S2 Ln 17:949 2Ln 4:2353 2:8875:
The normalized voltage values
12 2:5 0:8944;
V
2:795

11
V

1
0:3577:
2:795

Voltage change (10.27)


V121

0:8944  0:3577
0:7892:
1  0:8944  0:3577

We check expression (10.28)


V121

2  0:4888
0:7892:
1 0:48882

10.3.2 Case of Two Loads


Plane 
n1 
n2
Let the load voltages be equal to V11 = 1, V21 = 2.5, and parameters k12 = 1, k22 = 2.
The point C1n corresponds to the values V11 = 1, k12 = 1. Then, Eq. (10.29) has the
view
" 
#
5 2
0:5 0:8 n1 2  50:5 1n1 0:5 1 0:
21
From here, nC1
1 = 0.2269.
In accordance with (10.30)
nC1
2 1  2:5  0:2269 0:4326:
Similarly, we calculate the coordinates of the point D1n, which corresponds to
the values V11 = 1, k22 = 2.

10.3

Example

309

Table 10.1 Coordinates of


four points A1n, C1n, D1n, F1n
in the plane n1 n2

A1n
C1n
D1n
F1n

n1

n2


n1


n2

0.4
0.2269
0.26
0.4

1.3206
0.4326
0.7
1.3206

0.3578
0.203
0.2323
0.3578

0.9338
0.3059
0.4954
0.9338

The coordinates of the point F1n by (10.31)


nF1
1

0:4;

nF1
2

r
2
0:42 1:3206:
 2
1:25

The coordinates of all these points are presented in Table 10.1.


Moving of points along the line L1
Cross-ratio (10.32)
m21
nL1

0:2323 j0:4954  0:4  j0:9338


0:2323 j0:4954  0:4 j0:9338
0:203 j0:3059  0:4  j0:9338

1:6282:
0:203 j0:3059  0:4 j0:9338

We obtain the same cross-ratio for the points C2n, D2n with the voltage V21.
The coordinates of all required points are presented in Table 10.2.
Using (10.34), we dene
1 m21
1 1:6282
nL1
4:1837:

1:6282
1
m21

1
nL1
Then, the subsequent value

nD1

0:203 j0:3059  0:3578 j0:9338  4:1837  1


0:203 j0:3059  0:3578  j0:9338  4:1837

0:2323 j0:4953:

Table 10.2 Coordinates of


four points A2n, C2n, D2n, F2n
in the plane n1 n2

A2n
C2n
D2n
F2n

n1

n2


n1


n2

1.0
0.7325
0.7948
1.0

0.6324
0.2674
0.4104
0.6324

0.8944
0.6552
0.71088
0.8944

0.4472
0.1891
0.2902
0.4472

310

10

Regulation of Load Voltages

Moving of points along the line K1


Cross-ratio (10.35) or (10.36)
m21
nK1

0:6552 j0:1891 1 0:203 j0:3059 1



2:9135:
0:6552 j0:1891  1 0:203 j0:3059  1

Transformation ratio change (10.38)


n21
K1

2:9135  1
0:4889:
2:9135 1

Subsequent value (10.37)



nC2

0:203 j0:3059 0:4889


0:6552 j0:1891:
1 0:203 j0:3059  0:4889

1 V
2
Plane V
Moving of points along the line L1
11 0:3577 are given in
The coordinates of required points for the voltage V
Table 10.3.
Cross-ratio (10.41)
m21
VL1

0:7624 0:9338 0:5391 0:9338



2:6519:
0:7624  0:9338 0:5391  0:9338

Equality (10.42)
2:6519 1:62822 :
Voltage change (10.44)
V 21
L1

Table 10.3 Coordinates of


four points A1V, C1V, D1V, F1V
2 for the
1 V
in the plane V
11 0:3577
given V

A1V
C1V
D1V
F1V

2:6519  1
0:4523:
2:6519 1

n1

n2

V2

2
V

0.4
0.2269
0.26
0.4

1.3206
0.4326
0.7
1.3206

3.3015
1.906
2.6956
3.3015

0.9338
0.5391
0.7624
0.9338

10.3

Example

311

Subsequent value (10.43)


0:5391
 D1
V
2
0:9338 0:4523
0:8164:

F1

V2
1 0:5391
0:9338 0:4523

Moving of points along the line K1


11 0:3577, V
12 0:8944
Cross-ratio (10.45) for the voltages V
m21
VK1

0:8944 1 0:3577 1

8:4867:
0:8944  1 0:3577  1

21 2
Let us check the equality m21
VK1 = (mnK1) . Then

8:4867 2:91352 :
Voltage change (10.47)
V 21
K1

8:4867  1
0:7892:
1 8:4867

Subsequent value (10.46)


 21 0:3577 0:7892 0:8944:
V
1 0:3577  0:7892
 21 0:8944. The coordinates
Now, we use points of the line L1 for the voltage V
of required points are given in Table 10.4.
We check expression (10.49) for the points C1V, C2V, which correspond to
 C2
 C1
voltages V
2 ; V2 ,
q
2C1
V
21 2
1  VK1
21
11 VK1
1V
p
0:5391

1  0:78922 0:2582:
1 0:3577  0:7892

2C2
V

Table 10.4 Coordinates of


four points A2V, C2V, D2V, F2V
2 for the
1 V
in the plane V
 21 0:8944
given V

A2V
C2V
D2V
F2V

n1

n2

V2

2
V

1.0
0.27325
0.7948
1.0

0.6324
0.2674
0.4104
0.6324

1.5811
0.9128
1.2909
1.5811

0.4472
0.2582
0.3651
0.4472

312

10

Regulation of Load Voltages

In turn, expression (10.49) for the points D1V, D2V


q
2D1
V
21 2
1  VK1
21
11 VK1
1V
p
0:7624
1  0:78922 0:3651:

1 0:3577  0:7892

2D2
V

We see that the results of this calculation coincide with the data of Table 10.4.

References
1. Chae, S., Hyun, B., Agarwal, P., Kim, W., Cho, B.: Digital predictive feed-forward controller
for a DCDC converter in plasma display panel. IEEE Trans. Power Electron. 23(2), 627634
(2008)
2. Chattopadhyay, S., Das, S.: A digital current-mode control technique for DCDC converters.
IEEE Trans. Power Electron. 21(6), 17181726 (2006)
3. Chen, J., Prodic, A., Erickson, R.W., Maksimovic, D.: Predictive digital current programmed
control. IEEE Trans. Power Electron. 18(1), 411419 (2003)
4. Conformal geometry. Encyclopedia Wikipedia. http://en.wikipedia.org/wiki/Conformal_
geometry (2014). Accessed 30 Nov 2014
5. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw
Hill, New York (1967)
6. Glagolev, N.A.: Proektivnaia geometria. (Projective geometry). Nauka, Moskva (1963)
7. Hyperbolic geometry. Encyclopedia Wikipedia. http://en.wikipedia.org/wiki/Hyperbolic_
geometry (2014). Accessed 30 Nov 2014
8. Lui, X., Wang, P., Loh, P.: A Hybrid AC/ DC micro grid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
9. Penin, A.: Geometrical properties of regulated voltage converters at limited capacity voltage
sources. Tehnicheskaia elektrodinamika 3, 6164 (1991)
10. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
11. Penin, A.: Analysis of regimes of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Syst. 12(1), 12 (2013). http://www.wseas.org/
wseas/cms.action?id=6933. Accessed 30 Nov 2014
12. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013) https://sites.
google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 Nov 2014
13. Penin, A.: Non-Euclidean geometry and regulated characteristics of limited capacity power
supply. J. Electr. Eng. 2(4), 175186 (2014)
14. Stereographic
projection.
Encyclopedia
Wikipedia.
http://en.wikipedia.org/wiki/
Stereographic_projection (2014). Accessed 30 Nov 2014
15. Yehia, D.M., Yokomizu, Y., Iioka, D., Matsumura, T.: Delivaraible-power dependence on
distribution-line resistance and number of loads in low-voltage DC distribution system. IEEJ
Trans. Electr. Electron. Eng. 7(1), 2330 (2012)

Chapter 11

Stabilization of Load Voltages

11.1

Analysis of Load Voltage Stabilization Regimes

We use the same power supply system with two idealized voltage regulators
VR1; VR2, and load resistances R1, R2 in Fig. 11.1 [5, 6]. The regulators dene the
transformation ratios n1 = V1/V2, n2 = V2/V; an internal resistance Ri determines the
interference of the voltage regulators on load regimes.
For convenience, we rewrite Eq. (10.2)


Ri
Ri
V0 2 V0 2
2
2
:
V1
V2 V 

R1
R2
2
4

11:1

11.1.1 Case of One Load


In this case, the transformation ratio n2 = 0. Then,


Ri
V0 2 V0 2
2
:
V1 V 

R1
2
4

11:2

For different load values (R11 ; R21 and so on), this expression represents a bunch of
circles (ellipses) by the coordinates V1, V in Fig. 11.2.
Let the stabilized load voltage V1 = V1= be given [7]. Then, the vertical line with
the coordinate V1= intersects the bunch of circles in two points.
At the same time, the transformation ratio n1 is resulted by a stereographic
projection of circles points from the pole 0,0. Therefore, the load resistance R11
corresponds to the variable n11 ; R21 corresponds to n21 and so on.

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_11

313

314

11

Fig. 11.1 Power supply


system with two voltage
regulators VR1; VR2, and
loads R1 ; R2

Stabilization of Load Voltages

VR1
Ri

V1

V0

n1

VR2

V2
R

n2
Fig. 11.2 Stereographic
projection of bunch of ellipses
on a tangent line n1

V
n1

-R1
3

V0

n1

R1=

n1max

n1
1

n1

n1

R1min
2

R1
V0

R1

V1max
0
-R1

V1

V1=

For a minimum value R1min of the load resistance, the circle is tangent to the
vertical line V1=. In this case, V = V0/2. Using (11.2), we get the condition
Ri
R1min

V1 2

V0 2
:
4

11.1

Analysis of Load Voltage Stabilization Regimes

315

Then, the minimum load resistance value


R1min 4Ri

V1 2
V0 2

11:3

The respective maximum allowable transformation ratio


n1max 2

V1
:
V0

11:4

The operating area of all the circles must be above the diameters of these circles;
that is, V  V0 =2. Therefore, we use the upper point of the intersection.
In turn, on some step of switching period at increase of the parameter n1, a
running point may pass over the diameter that is inadmissible. Therefore, it is better
to use such groups of transformations or movements of points along the line n1
when it is impossible to move out the running point over the diameter. So, we must
decrease the next values n1 by some rule. In this sense, we come to hyperbolic
geometry.

11.1.2 Use of Hyperbolic Geometry


Let us consider the dependence R1 n1 ; V1 , where the voltage V1 is a parameter.
Similarly to (10.15) and (10.17), we must validate the denition of regime and its
changes and nd the invariants of regime parameters. To do this, we consider such
a characteristic regime as R1 1. In this case, the ellipse degenerates into the two
straight lines, V = 0, V = V0.
Then, for the voltage V = V0, the transformation ratio
n11

V1
:
V0

11:5

However, the question arises about the range 0\n1 \n11 of this transformation
ratio. According to Fig. 11.2, this range corresponds to negative load values R1
and expression (11.2) determines a hyperbola. In this case, the load gives energy
back and the voltage source V0 consumes this energy as it is shown in Fig. 11.3.
In this regard, we consider a physical realization of such a power source, as a
negative resistance. For this purpose, we remind something of the electric circuit
theory by examples of two circuits in Fig. 11.4.
Let a voltage source V1 be connected to a resistance R1 for the left-hand circuit.
Then, the negative resistance R1 and the drain current I1 correspond to this
voltage source.
We consider Fig. 11.3 again. Then, it is possible to connect up the voltage source
V1 instead of the resistance R1 . At the same time, the input resistance of this

316

11

Fig. 11.3 Negative load R1


gives energy back

Ri

Stabilization of Load Voltages

I1

VR1

V0

I0

V1

-R1
-

n1
Fig. 11.4 Negative resistance
R1 corresponds to a voltage
source V1 and its practical
realization by PWM boost
converter

I1

I0

I1

R1
V1

Ri

V1

V0
-R1

R1

circuit must be equal to the constant value R1 at the voltage V1 change. This
condition is satised due to the variable value n1. We obtain so-called a loss-free
resistance [8]. The example of another such a circuit is the right-hand circuit in
Fig. 11.4 as a high-power-factor boost rectier [4].
Taking into account the value n1 = V1=/V and Eq. (11.2), we obtain
V1

n1 V0
1

Ri
R1

n1 2

11:6

Thus, the required relationship n1 R1 and its inverse R1 n1 is obtained


n1 2  n1

V0 R1 R1

0;
V1 Ri
Ri

R1
n1 2

:
Ri n1 VV0  1
1

11:7

In turn, the dependence R1 n1 determines a hyperbola in Fig. 11.5.


We have a single-valued mapping of hyperbola points onto the axis n1. This
projective transformation preserves a cross ratio of four points. Similarly to (10.15),
let us constitute the cross ratio m1n for the points 0; n11 ; n11 ; n1max ; that is,
m1n 0 n11 n11 n1max

n11  0
n11  0

;
 n1max n11  n1max

n11

where the points 0; n1max are base ones and n11 is a unit point.

11:8

11.1

Analysis of Load Voltage Stabilization Regimes

317

Fig. 11.5 Dependence


R1 n1

R1

R1min

n1

n1max

n1

Using (11.4) and (11.5), we get



m1n

0 n11

V1
V0

V1
V0

n11
n11

:
2 VV10  n11 n1max  n11

11:9

The conformity of the points n11 ; m1n is shown in Fig. 11.6. In this case, the value
m1n is a non-homogeneous coordinate of n11 .
1
2
Further, the cross ratio m21
n , which corresponds to the regime change n1 ! n1 ,
has the form
2 1
m21
n 0 n1 n1 n1max

n21 n11  n1max


:
n11 n21  n1max

11:10

Using the normalized values


n21

n21
n1max

n11

n11
n1max

Fig. 11.6 Conformity of


different regime parameters

21

n1
0

n1 n1
1

n1max

n1

mn

R1

R1min R1

I1max

n1
21

mn
0

mn

mn
21

R1
0

R1

21

I1

I1

I1

I1

318

11

Stabilization of Load Voltages

we get the regime change (segment n21 n11 ) or cross ratio (11.10) in the view
21 n11 1
m21
n 0 n

n21 n11  1
:
n11 n21  1

11:11

Similarly to (10.19), we may obtain the analogous expression for the change n21
1
of the transformation ratio so that the following relationships are performed
n21
1

m21
n 1
;
m21
n 1

m21
n

n21
1 1
:
1  n21
1

11:12

For this purpose, we make the substitution of variables so that to use ready
expressions (10.19). Therefore, we introduce the value
~n1 2n1  1

11:13

This simple conformity of variables is shown in Fig. 11.7.


According to (10.21),
~n21
1

~n21  ~n11
:
1  ~n21 ~n11

Using substitution of variables (11.13), we get


~n21
1

n21

n21  n11
n21
1 :
n11  2n21 n11

11:14

In this expression, the changes of the variables are equal to among themselves;
21
that is, ~
n21
1 n1 .
There is a following foundation for this equality. Linear expression (11.13)
preserves a cross ratio and, consequently, the regime change m21
n . On the other
21
hand, the change of the transformation ratio n1 is expressed by regime change
(11.12).

Fig. 11.7 Conformity of the


normalized transformation
n1
ratio n1 and a new variable ~

0.5

n1

-1

n1

n1

n1

11.1

Analysis of Load Voltage Stabilization Regimes

319

Using (11.14), we obtain the subsequent transformation ratio


n21

n11 1 n21
1
:
21
1
1 n1 2n1  1

11:15

There is a group transformation. Moreover, if the initial value n11 1, then the
subsequent value n21 1 regardless of the value n21
1 . Therefore, such movement of
point corresponds to hyperbolic geometry.
Similarly to the above, let us consider the cross ratio for the load resistance R1 .
Using the dependence R1 n1 in Fig. 11.5, we demonstrate the conformity of the
variables R1 ; n1 by Fig. 11.6. The cross ratio for the initial point R11 relatively to the
base points 0; R1min , and a unit point R1 1 has the form
m1R 0 R11 1 R1min

R11  0
1

:
 R1min 1  4 R1i V1 22
R V

R11

11:16

Expression (11.16) equals the corresponding cross ratio for the conductance Y1
1=R1 and the load current I1 U1 =R1 . Also, the following equality takes place
mR mn 2 :

11:17

This expression leads to identical values if we use the hyperbolic metric to


determine a regime value or distance
S Ln mR 2Ln mn :

11:18

The base points 0; R1min correspond to the innitely large distance.


Similarly to (11.10), the cross ratio m21
R , which corresponds to the regime change
R11 ! R21 , has the view
2
1
m21
R 0 R1 R1 R1min

R21 R11  R1min


:
R11 R21  R1min

11:19

We may introduce the change R21


1 of the load resistance by the following
expression
m21
R

1 R21
1
:
1  R21
1

Further, we use the normalized values


2
 21 R1 ;
R
R1min

1
 11 R1 :
R
R1min

11:20

320

11

Stabilization of Load Voltages

Similarly to (11.14), we get


 11
 21  R
R
R21
1 2
 11 :
 11  2R
 21 R
R1 R

11:21

Then, there is a strong reason to introduce the value of load resistance change
21
as R21
1 and the value of transformation ratio change as n1 .
The validity of such denitions for changes is conrmed by the following
expression similar to initial expression (11.6); that is,
R21
1

2n21
1
2
1 n21
1

11:22

Using (2.21), we obtain the subsequent value R21 of the load resistance
 21
R

 11 1 R21
R
1
:
1  1

1 R21
2
R
1
1

11:23

 21 1
 11 1, then the subsequent value R
As well as (11.15), if the initial value R
regardless of the value R21
1 .
Thus, a concrete kind of a circuit and character of regime imposes the
requirements to denition of already system parameters.
Therefore, arbitrary and formal expressions for regime parameters are excluded.
Example Let the circuit parameters be given as follows
V0 5;

Ri 1;

V1 2:5:

The initial and subsequent value of the load resistance


R11 2:0;
Minimum load resistance (11.3)
R1min 4Ri

R21 1:25:

V1 2
V0 2

1:

Maximum transformation ratio (11.4)


n1max 2

V1
1:
V0

11.1

Analysis of Load Voltage Stabilization Regimes

321

Transformation ratio (11.5)


n11

V1
0:5:
V0

The values of the transformation ratio by quadratic Eq. (11.7)


n11 0:585;

n21 0:691:

The normalized values are equal to these transformation ratios.


Cross ratio (11.8) for the initial regime
m1n 0 n11 n11 n1max

0:585
1:41:
1  0:585

Cross ratio (11.10) for the regime change


1
m21
n 0 n1 n1max

n21 n11  n1max 2:236


1:581:

1:41
n11 n21  n1max

Change (11.14) of the transformation ratio


n21
1

n21  n11
0:106
0:226:

2
1
1
2
n1 n1  2n1 n1 0:467

Now, we consider the cross ratio for the load resistance R1 .


Cross ratio (11.16) for the initial regime
m1R 0 R11 1 R1min

1
1

2
4 RRi1VV12
1 0

1
2:0:
1  4 6:25
225

Let us check equality (11.17); that is,


m1R m1n 2 1:412 2:0:
Cross ratio (11.19) for the regime change
2
1
m21
R 0 R1 R1 R1min

R21 R11  R1min


2:5:
R11 R21  R1min

The equality
21 2
2
m21
R mn 1:581 2:5:

322

11

Stabilization of Load Voltages

Change (11.21) of the load resistance


 11
 21  R
R
R21
1 2
 21 0:429:
 11  2R
 11 R
R1 R
Equality (11.22)
R21
1

2n21
1
2
1 n21
1

2  0:226
0:429:
1 0:051

11.1.3 Case of Two Loads


Let us now consider a circuit with two loads in Fig. 11.1. The variation of one of
these loads leads to a mutual change of stabilization regimes for both loads.
Therefore, it is necessary to change the transformation ratios n1, n2 in some
coordination for the stabilization of V1 ; V2 . We will obtain the required
relationships.
We rewrite Eq. (11.1)


Ri
Ri
V0 2 V0 2
:
V1 2
V2 2 V 

R1
R2
2
4

11:24

By denition,
n1

V1
;
V

n2

V2
:
V

11:25

Using (11.6), (11.24) and (11.25), we get the system of equations


8
< V1 1 Ri nn12V0 Ri n
: V2

2
R2 2
n2 V0
R
R
1 R i n1 2 R i n2 2
R1

11:26

It follows that
(

Ri
R1
Ri
R1

n1 2
n1 2

Ri
R2
Ri
R2

V0
n2 2  nV11
1 0
2
n2 V0
n2  V2 1 0:

11:27

11.1

Analysis of Load Voltage Stabilization Regimes

323

By denition (11.25)
n2 n1

V2
:
V1

11:28

Substituting this expression in the rst equation of system (11.27), we get


"


 #
1
1 V2 2
n1 V 0
n1 Ri

1 0:

R1 R2 V1
V1
2

11:29

The expression in the square brackets is the total resistance (conductance) of


both loads relatively to the rst load; that is,


1
1
1 V2 2

YT :
RT R1 R2 V1

11:30

If, for example, the load voltages are equal to among themselves, V2 V1 ,
then these loads are connected in parallel and
1
1
1

YT :
RT R1 R2
Finally, we get the expression
n1 2

Ri n1 V0

1 0:
RT
V1

11:31

This expression corresponds to (11.7) and the dependence RT n1 coincides with


Fig. 11.5.
Therefore, for given load resistances, we determine:
total resistance (11.30);
transformation ratio n1 as the solution of (11.31);
value n1 by (11.28) or by the second equation of (11.27).
Also, we must check stability conditions (11.3), (11.4). In this case, these
conditions have the form
(

RTmin 4Ri VV12


n1max 2 VV10 ;

n2max 2 VV20 :

11:32

Further, it is possible to use the above idea of hyperbolic geometry in the case of
one load.

324

11.2

11

Stabilization of Load Voltages

Given Voltage for the First Variable Load


and Voltage Regulation of the Second Given Load

We consider again the circuit shown in Fig. 11.1. Let the rst load voltage V1 be
stabilized. But for all that, the rst load resistance may be both positive R1 [ 0 and
negative R1 \0. Also, the second constant load resistance is positive R2 [ 0.
For example, the circuit in Fig. 11.8 corresponds to the positive load R1 [ 0 and
PWM regulators in Fig. 11.9a conform to the negative load R1 \0.
We rewrite Eq. (11.1)


Ri
Ri
V0 2 V0 2
2
2
;
V1
V2 V 

R1
R2
2
4
which correspond to a surface with a parameter R1 in the coordinates V1, V2, V3.
If R1 [ 0, this expression represents a sphere (ellipsoid) similarly to the circle in
Fig. 11.2. The both loads consume energy; the voltage source V0 gives energy.
If R1 \0, we get a one-sheeted hyperboloid in Fig. 11.9b [3]. The rst load, as a
constant voltage source V1 , gives energy. In addition, the voltage source V0, as
energy storage, may consume and give back energy. The corresponding direction of
the current I0 determines these regimes.
For different values R1, our expression represents a bunch of spheres or
hyperboloids. If R1 1, as the open circuit regime, the corresponding surface
degenerates into a cylinder.

VR1
Ri
V0

I1

V1=

I0

VR2

n1

I2
V2
R

n2
Fig. 11.8 Power supply system with the invariable voltage V1 and given load R2

11.2

Given Voltage for the First Variable Load

325

I1

(a)

I2
Ri

V1=

V2

I0

V0

R2

R1

(b)

V
V0

V2

0
V1
Fig. 11.9 a Power supply system with resistance R1 . b Its geometric model

Let us now return to given operating regime; that is,


V1 V1 ;

R2 const:

For realization of this regime, it is necessary to change the transformation ratios


n1 ; n2 in some coordination. Let us obtain the required relationships.
So, the plane with the parameter V1 intersects the bunch of spheres and
hyperboloids. As the result of this section, the bunch of circles in coordinates V2, V
is obtained, as it is shown in Fig. 11.10.
In this case, our expression has the form


Ri
V0 2 V0 2 Ri
 V1 2 :
V2 2 V 

R2
2
4
R1

11:33

The second member of this equation is a radius of circle for the given value R1.
It is possible to consider the voltage V2 change as the radius-vector rotation. This
rotation determines a point movement along the straight line V1 in coordinates V1,
V in Fig. 11.11. This gure at V2 = 0 is analogous to Fig. 11.2.

326

11

Stabilization of Load Voltages

Fig. 11.10 Bunch of circles


for different R1 as V1

V
V0

R1=

-R1

R1
V0
2

Fig. 11.11 Bunch of curves


at V2 0 and a point moving
along lines V1

V2

V
-R1
V0

n1

n1max

R1 =

n1
V0
2
R1>0
V1
0

-R1

V1=

In the general case of the variable V2, we get the surfaces, which rotate around
the diameter V = V0/2, as it is shown by closed arrows. Also, the transformation
ratios n1, n2 are resulted by the familiar stereographic projection of spheres points
on the tangent plane or conformal plane. The axes n1, n2 are superposed in
Fig. 11.11.
Further, we use the rst equation of system (11.27)
Ri
Ri
n1 V0
n1 2
n2 2 
1 0:
R1
R2
V1

11:34

11.2

Given Voltage for the First Variable Load

327

This equation circumscribes trajectories for different values R1, as it is shown in


Fig. 11.12. These trajectories are characteristic lines for the conformal plane.
If R1 > 0, the bunch of circles with parameters R11 ; R21 is obtained.
For R1 1, Eq. (11.34) corresponds to a parabola; that is,
Ri
n1 V0
n2 2 
1 0:
R2
V1

11:35

The case R1 \0 conforms to a hyperbola. For the limit values R1 0 and


n1 0, the hyperbola degenerates and coincides with the axis n2.
The radius-vector rotation in the plane V2V determines the analogous rotation in
the plane n1n2 with the center n1max. This center corresponds to the minimum load
resistance R1min.
Let us nd these values n1max, R1min. We assume n2 = 0 in expression (11.34).
Then, we get the quadratic equation
Ri
n1 V0
n1 2 
1 0:
R1
V1

11:36

The roots coincide for


R1 R1min 4Ri

V1 2
V0 2

Using (11.36), we get


n1max 2

V1
;
V0

These values equal (11.3) and (11.4) for one load.

Fig. 11.12 Trajectories for


different values R1 as V1

n2

R1<0

R1=
2

R1
1

R1
0

n1max

n1

328

11

Stabilization of Load Voltages

Next, we display the voltage V2 value for the trajectories in Fig. 11.12. Using
denition (11.25), we obtain
V2 V1

n2
:
n1

Therefore, the voltage V2 is directly proportional to the voltage V1= as constant


value n2/n1; that is, we have a straight line, which intersects the bunch of circles
with the parameter R1 in two points. Then, the tangent lines to the circles (curves)
determine the points of the maximum voltage V2max, the voltage V = V0/2.
In this case, we get
V2max

V0
2

rr
R2
R1min
1
;
Ri
R1

n2max

rr
R2
R1min
1
:
Ri
R1

11:37

It is interestingly to note that all these tangent lines to the curves correspond to
the value n1max . Therefore, the operating area of the transformation ratio is limited
by the value n1  n1max . So, we must decrease the next value n1, for the next
regulator switching period, by some rule. In this sense, we come to hyperbolic
geometry.

11.2.1 Use of Hyperbolic Geometry


We may suppose that the straight line n1max in the plane n2 n1 is the innitely
remote line. Therefore, geometry of the half plane n2 ; n1  n1max in Fig. 11.12 and
of normalized half plane n2 n1  1 in Fig. 11.13 corresponds to Poincares model of
hyperbolic geometry. This Poincares model is also demonstrated by projective
coordinates g2 ; g1 for the left-hand gure and by these Cartesian coordinates for the
right-hand gure in Fig. 11.13 for the half plane g2 ; g1  0.
For Poincares model of hyperbolic geometry, the half-rounds with the given
resistance R1 intersect the axes g2 orthogonally. Let us introduce this geometry. To
do this, it is necessary to change the variables n2 g1 ; g2 ; n1 g1 ; g2 so that all
curves of the plane n2 n1 would be converted into circles of the plane g2 g1 .
Further, we use the normalized values
n1

n1
n1max

n2

n2
:
n2ref

As the scale value n2ref , we may use a circle with some characteristic value of the
parameter R1. The resistance value R1 1 may be such a characteristic value.
Using (11.35) and the value n1max , we get

11.2

Given Voltage for the First Variable Load

Fig. 11.13 a Poincares


model of hyperbolic geometry
for superposed half planes
n2 ; n1  1 and g2 ; g1  0.

b Half plane for orthogonal
coordinates g2 ; g1

329

(b)

(a)

g2

n2

g2

R1= 1

R1<0
R1
0

0.5
1

R1

n1

g1

R1= 1

g1

-1

R1<0

n2ref

r
R2
:

Ri

11:38

It is possible to represent expression (11.34) in the normalized form


R1min
n1 2 n2 2  2n1 1 0:
R1

11:39

The required change of variables has the view [1]


n1

1
;
1 g1

n2

g2
:
1 g1

11:40

Let us check the offered expressions. In this case, Eq. (11.39) transforms into the
equation of circle; that is,
g1 2 g2 2 1 

R1min
r1 2 :
R1

11:41

The second member of this equation is a radius squared for the given R1. We
may term the variables g1 ; g2 as hyperbolic transformation ratios.
This geometric model allows to use a cross ratio for the determination of regimes
and their change.

330

11

Stabilization of Load Voltages

11.2.2 Regime Change for the First Given Load Resistance


For clarity, let us consider the half-rounds with the parameter R11 in Fig. 11.14.
Let points C1g ; D1g be points of an initial and subsequent regime. Then, the cross
ratio, which corresponds to the regime change C1g ! D1g , has the form similar to
(11.10); that is,
mDC
g A1g D1g C1g F1g

D1g  A1g C1g  A1g



:
D1g  F1g C1g  F1g

11:42

The points A1g ; F1g are the base points.


The coordinates of all the points A1g ; D1g ; C1g ; F1g are given by complex numbers as follows
gA1 gA1
2 j0;
C1
gC1 gC1
2 jg1 ;

D1
gD1 gD1
2 jg1 ;

gF1 gF1
2 j0:

F1
In particular, radius of half-rounds (11.41) denes the coordinates gA1
2 ; g2 as the
follows

gA1
2 r1 ;

gF1
2 r1 :

Fig. 11.14 Regime change


for Poincares model of
hyperbolic geometry

g2

D2g

1
F1g

D1g

C2g

C1g

jg1

1 B2g

B1g
1

R1
A1g

-1

A2g

11.2

Given Voltage for the First Variable Load

331

For Poincares model of hyperbolic geometry by Fig. 11.14, cross ratio (11.42)
looks like [2]
mDC
g

tghC r1  gC1
r1  gD1
2
2


:
D1
tghD
g
gC1
1
1

11:43

Using (11.41), we get




mDC
g

2

r1  gC1
r1  gD1
2
2

:
C1
r1 gD1
r1 g2
2

11:44

This expression gives the subsequent value


gC1

2
g2
gD1
2
r1 gC1
;
r1
1 r21 gDC
2

DC

where the hyperbolic transformation ratio change is introduced as



2
mDC
1
g
:

gDC


2
2
mDC

1
g

11:45

11:46

This change corresponds to the points C2g ; D2g of the half-rounds with parameter
R21 and so on.
We may obtain an expression for the subsequent value of the transformation
ratios n1 ; n2 . To do this, it is necessary to apply to (11.45) the inverse change of
variables relatively to (11.40); that is,
g1

1  n1
;
n1

g2

n2
:
n1

11:47

But complicated formulas are obtained. Therefore, using (11.40), we may


directly calculate the subsequent value of transformation ratios n1 ; n2 .
Let us now compare the half-rounds in the plane g2 g1 with the half-rounds in the
plane V2 ; V in Fig. 11.15.
The points A1V ; F1V are the base points. The points C1V ; D1V correspond to the
initial and subsequent regime. It is possible to conclude that the half plane
V2 ; V  V0 =2 is also hyperbolic geometry model.
Therefore, similarly to (11.44), the regime change has the view
D1
mDC
V2C1 V2max
V V2max V2

V2D1 V2max V2C1 V2max



:
V2D1  V2max V2C1  V2max

11:48

332

11

Fig. 11.15 Regime change


for hyperbolic geometry in the
half plane V2 ; V  V0 =2

V
B2V

Stabilization of Load Voltages

V0

B1V

C1V

R1

A2V

C2V

D1V

V0
2

A1V

D2V

F1V

V2

Also, the following equality takes place



2
DC
mDC
:
V mg
Using (11.48), we obtain the subsequent voltage
V C1

2
V2DC
V2D1
V2max V C1
;
V2max 1 2 V DC
V2max 2

11:49

where the voltage change is introduced as


V2DC

mDC
V 1
gDC
2 :
mDC
V 1

11:50

It now follows that


V2D1
gD1
2 :
V2max
r1
The obtained expressions may be generalized for three or more loads.

11:51

11.2

Given Voltage for the First Variable Load

333

11.2.3 Example
Let the circuit parameters be given as follows
V0 5;

Ri 1;

R2 2;

V1 2:5;

R1min 1;

n1max 1:

Scale value (11.38), n2ref 1:414.


The initial regime, the point C1, is set by the rst load resistance R11 1:25 and
the second load voltage V2C1 0:707.
Using (11.24), we get

V

3:5
:
1:5

Further, we use the voltage value V C1 3:5 because this value is greater than
V0 =2 2:5.
Maximum values (11.37)
V2max
n2max

r
pp
5 p
1
2 1

2:5 2 0:2 1:581;


2
1:25
pp
2 0:2 0:6324:

p
Radius (11.41), r1 0:2 0:447.
Transformation ratios (11.25)
nC1
1

2:5
0:714;
3:5

nC1
2

0:707
0:202:
3:5

The normalized transformation ratios


nC1
1 0:714;

nC1
2

0:202
0:143:
1:414

Hyperbolic transformation ratios (11.47)


gC1
1

1  0:714
0:4;
0:714

gC1
2

0:143
0:2:
0:714

We check expression (11.41)


0:42 0:22 0:2:
Further, we are verifying the regime change for the given load resistance R1.
Let the subsequent regime, the point D1, be given by the second load voltage
V2D1 1:414 and voltage V D1 3.

334

11

Stabilization of Load Voltages

Therefore,
2:5
1:414
0:833; nD1
0:333;
2
3
3  1:414
1  0:833
0:333
0:2; gD1
0:4:

2
0:833
0:833

nD1
1
gD1
1

Regime changes (11.43), (11.44)


p
p
0:2  0:2
0:2  0:4

2:618:

0:4
0:2
p


2
0:2  0:2
0:2  0:4
mDC
 p
6:8528 2:6182 :
p
g
0:2 0:2
0:2 0:4
mDC
g

Transformation ratio change (11.46)


gDC
2

6:8528  1
0:7453:
6:8528 1

Subsequent value (11.45)


p0:2
0:7453
gD1
0:4
2
0:20:2
0:8944 p :
 0:7453
r1
1 p
0:2
0:2

Regime change (11.48)


mDC
V

1:414 1:581 0:707 1:581



6:8528 2:6182 :
1:414  1:581 0:707  1:581

Equality (11.51)
1:414
0:4

0:8944:
1:581 0:4472

References
1. Glagolev, N.A.: Proektivnaia geometria. (projective geometry). Nauka, Moskva (1963)
2. Kagan, V.F.: Osnovania geometrii, Chasti II. (Geometry basics. Part II). Gostekhizdat, Moskva
(1956)
3. Korn, G.A., Korn, T.M.: Mathematical handbook for scientists and engineers. MacGrawHill,
New York (1968)
4. Maksimovic, D., Jang, Y., Erickson, R.W.: Nonlinear-carrier control for highpowerfactor
boost rectiers. IEEE Trans. Power Electron. 11(4), 578584 (1996)

References

335

5. Penin, A.: Analysis of regime of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Syst. 12(1), 12 (2013). http://www.wseas.org/
wseas/cms.action?id=6933 Accessed 30 Nov 2014
6. Penin, A.: Non-Euclidean geometry and regulated characteristics of limited capacity power
supply. J. Electr. Eng. 2(4), 175186 (2014)
7. Penin, A.: NonEuclidean geometrical transformation groups in the electric circuit theory with
stabilization and regulation of load voltages. Int. J. Circuits Syst. Sign. Process. 8, 182194
(2014). http://www.naun.org/cms.action?id=7621 Accessed 30 Nov 2014
8. Singer, S., Erickson, R.W.: Powersource element and its properties. IEE ProceedingsCircuits,
Devices and Systems. 141(3), 220226 (1994). http://ecee.colorado.edu/*rwe/papers/IEE94.
pdf

Chapter 12

Pulse-Width Modulation Regulators

12.1

Introduction

As it was noted in Chap. 10, into power supply systems with limited capacity
voltage sources, the limitation of load voltage is appeared. Renewable power supply
with solar array, fuel cell, and rechargeable battery may be the example of ones [4].
The output voltage of these sources changes over a wide range. In such systems it is
convenient to apply the pulse-width modulation PWM boost and buckboost regulators or converters which can step up or step down the input voltage [2, 10], but
these regulators have a nonlinear regulation curve or characteristic. Therefore, the
problem of linearization is appeared [9].
The linearization methods are proposed for idealized, without losses buckboost
regulators [3, 5].
Also, the real boost and buckboost converters have the two-valued regulation
characteristic. In this case, the up-slope direction or the forward branch of its
regulation characteristic is used and the down movement of the operating point on
the back branch is restrained.
Therefore, it is necessary to correctly determine the regime parameters of the
converter relatively to the maximum permissible load voltage and control pulse
width. It will allow estimating, for example, reserves of the control voltage and load
voltage, to use this data for a direct digital or predictive control, and to carry out
some kind of the linearization of the regulation characteristic in a wide range of
load voltage changes.

12.2

Regulation Characteristic of Boost Converter

Let us consider a PWM boost voltage converter with a given load resistance RL in
Fig. 12.1 [8].
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_12

337

338

12

Fig. 12.1 Boost voltage


converter

Pulse-Width Modulation Regulators

VL

Ri

CL

RL

V0

0
PWM

Vreg

We know the expression of the static regulation characteristic for the continuous
current mode of the choke L with the loss resistance Ri [1]
VL V0

1D
1  D2

Ri
RL

V0

1
1D
r2
1 1D
2

12:1

where D  1 is a relative pulse width and r is a relative loss. For convenience, let
the converter be given by the following parameters:
V0 25;

r 0:08:

Then, characteristic (12.1) has the view of Fig. 12.2.

Fig. 12.2 Example of


regulation characteristic via
the pulse width

160

VL
120

80

40

0
0,0

0,2

0,4

0,6

0,8

1,0

12.2

Regulation Characteristic of Boost Converter

339

We may introduce a value


n

1
 1:
1D

12:2

This value is the inverse relative pause width or the transformation ratio of the
idealized boost converter
n

VL
:
V

12:3

Thus, we obtain that


VL V0

n
1 r n2

12:4

This equation corresponds to regulation characteristic (10.8) for power supply


system with one load of Sect. 10.2.1. Therefore, we may use the results of
this section directly. Regulation characteristic (12.4) for our example is given in
Fig. 12.3.
Expression of ellipse (10.2) obtains the general form
r2 VL 2 V2  V0 V 0:

12:5

The plot of this ellipse is given in Fig. 12.4.

Fig. 12.3 Example of


regulation characteristic via
the transformation ratio

160

VL
80

-80

nM
-160
-15

-10

nM

1
-5

10

15

340

12

Fig. 12.4 Geometrical model


of the regulation characteristic
of the boost converter

Pulse-Width Modulation Regulators

V
n

V0

nM

V0
2
0

VL VLM

VL

The variable n turns out at the expense of a stereographic projection too.


Therefore, the maximum values
VLM 

V0
156:25;
2r

1
nM  12:5:
r

12:6

coincide with (10.5) and (10.6).


We note that the real working area n  1; but the whole area from nM to nM
allows using the above results.
We consider that a regime change or load voltage regulation is dened by a
group hyperbolic transformation, which consecutively, step by step, translates an
initial point VL1 into a subsequent point VL2 and so on.
The conformity of the characteristic points and running points of different
variables is shown in Fig. 12.5.
Such transformation possesses a familiar invariant; it is a cross ratio of four
points. We determine necessary points of characteristic regimes. There are two
points of the maximum voltage: VLM . Also, there is a unit or starting point, when
VL0

V0
1 r2

n 1;

D 0:

12:7

For the initial values n1 ; VL1 , the corresponding cross ratios are
m1n


 1
n1 1  r
1
1
;
 n1 1
:
r1
1 1r
r
r
rn


m1v


V0 1 V0 V0
V

2r L 1 r2 2r

V0
2r
V0
2r

VL1 1  r2
:
m1n 2 :
 VL1 1 r2

12:8

12:9

12.2

Regulation Characteristic of Boost Converter

Fig. 12.5 Regime change as


a group transformation and
conformity of the
characteristic and running
points of the variables
VL ; n; D

341

V0
V0 1

21

VL

1+

VL

VL

V0 1

VL

21

n
1

1+

D 1

n
D

21

mV
SCL

mV

mV

mV

mV
SCL

SCL

0
0

By denition (12.2), the values D and n are connected among themselves by the
fractionally linear expression
D

n1
:
n

Then, it is possible at once to express cross ratio (12.8) by the value D




m1D m1n 1 r D1 0 1 r

1 r  D1 1  r
:
:
1  r  D1 1 r

12:10

Expressions (12.8) and (12.9) lead to identical values if we take the logarithm
S1 Ln m1v 2Ln m1n :
Then, the hyperbolic distance
V0
1
VL1
1r
1r
2r VL

Ln
:

2Ln
 2Ln
V
1
1
0
1r
1r
2r  VL
2r  VL

V0

S1 Ln 2r
V0

342

12

Pulse-Width Modulation Regulators

In particular, for a unit point, this hyperbolic distance is equal to zero.


Besides the considered three points of the characteristic regimes, there is still a
fourth one, the scale points V 0; n 0; D 1. This scale point should be
considered too. The cross ratio and hyperbolic distance for the scale point are
mSCL
V



1r 2

\1;
1r

Ln mSCL
2Ln
V

1r
\0:
1r

The corresponding hyperbolic distance will be positive for inverse value of the
cross ratio:
SSCL 2Ln

1r
[ 0:
1r

Further, it is natural to introduce the normalized hyperbolic distance for a running regime (the index 1 is lowered), using the obtained scale
r

VL
1

 1:
r
2Ln 11r
2r  VL

V0

S
SSCL

Ln 2r
V0

12:11

Thus, the normalized distance considers all the characteristic points. In turn, the
inverse expression is
 r2r 1
1
V0 11r
:
VL
1 r2r 1
2r
1

12:12

1r

For a regime change n1 ! n2 , we have


1
n n
1
n2 r1 n1 r 1r2 n2 n1
1
1
 n2 n1 r1


:
1
1
n2 n1
2
1
r
r
rn
rn
r  1r2 n2 n1
2

m21
n

It is possible to introduce, similar to (10.21), the transformation ratio change as


n21

n2  n1
1  r2 n2 n1

12:13

Then,
n21
:
21
rn

r
m21
n 1

12:14

12.2

Regulation Characteristic of Boost Converter

343

The subsequent value of the transformation ratio is


n2

n1 n21
1 r2 n1 n21

12:15

Similarly, for the voltage change VL1 ! VL2 , it is possible to write down at once
VL21
;
21
2r  VL

V0

2r
m21
V V0

VL21

VL2  VL1
1  4r2

VL2 VL1
V0 2

VL2

VL21 VL1
1 4r2

VL21 VL1
V0 2

12:16

Let us suppose the change m21


D . Using (12.10), we have
2
1
m21
D mD  mD

1 r  D2 1 r  D1 1  rD21


;
1  r  D2 1  r  D1 1 rD21

where the change is


D21

D1  D2
;
1  r2 D1 D2  D1 D2

n21 D21 :

12:17

Thus, the mutually coordinated system of all the regime parameters turns out.
We will consider the linearization of the regulation characteristic of a converter
in Fig. 12.6 [6]

Fig. 12.6 Boost converter


with linearization function
calculators of the control
voltage DV and feedback
voltage Vfb

VL

Ri

RL

CL

inverse
nonlinear
function
calculator

V0
0

PWM

Vreg

Vfb
nonlinear
function
calculator

Vref
0

344

12

Pulse-Width Modulation Regulators

Let us express D and n via r similar to (12.12). Then


D1r

 1 r r 1

1
1r
;
 1 r r 1
1
1r

 r r 1
1
1 11r
n 
:

r

1
r 1r
1

12:18

1r

It turns out that the value r is equally expressed by D and VL. It is possible to
interpret this equality as the linearization of the dependence VL(D).
Further, it is possible to accept that the value D is equal to the regulating voltage
Vreg for PWM. We believe that the input voltage DV of a nonlinear function
calculator (introduced before PWM) is the hyperbolic distance r.
Therefore, we obtain, by (12.18), that
1 rDV 1

Vreg 1  r 1r DV 1


1r
1r

1
1

12:19

The plot of the function Vreg DV is given in Fig. 12.7.


Additionally, expressing r via VL by an inverse nonlinear function calculator
according to (12.11), we obtain a value close to the voltage DV, that is, the feedback voltage Vfb in Fig. 12.8. Therefore, we rewrite expression (12.11)
VL
1
 1:
r
2Ln 11r
2r  VL

V0

Vfb Ln 2r
V0

12:20

So, we obtain that Vfb is equal to DV. If to use a feedback closed loop (shown by
the dashed line in Fig. 12.6), the output voltage of the error amplier will be the
voltage DV. ORCAD model of this converter is proposed by [8].
Fig. 12.7 Nonlinear
characteristic of the function
calculator

1.0

Vreg
0.8

0.6

0.4

0.2

0.0
0

10

15

12.2

Regulation Characteristic of Boost Converter

Fig. 12.8 Nonlinear


characteristic of the inverse
function calculator

345

16

Vfb
12

0
0

40

80

120

VL

160

Examples We use the above converter parameters as V0 25, r 0:08.


The characteristic values of all the regime parameters are as follows:
V0
1
156:25; nM  12:5;
r
2r
nM  1
DM
1  r 0:92:
nM

VLM 

Example 1 Let VL1 48:49 be the initial regime. We must dene the normalized
distance for this regime.
For n 1; we have a unit value
VL 25

1
24:841:
1 0:082

All these values are shown in Fig. 12.9.


Cross ratio of initial regime (12.9) is
m1V

156:25 48:49
 0:8522 1:378:
156:25  48:49

Then,
m1n

q p
m1V 1:378 1:174:

346

12

Fig. 12.9 Example of a


regime change and
conformity of points of the
variables VL ; n; D

Pulse-Width Modulation Regulators

24.841 48.49
156.25 VL

69.85

156.25

2.949
12.5

1.986

12.5

0.92

0.661

1.08

0.496
1.9

0.7256

1.378

mV

0.32

0.32

The corresponding value n1 may be calculated from the inverse formula to (12.8)
n1

r 1
m 1
1 11r
 1 r n1
1:986:
r 1r mn 1

In turn,
D1 0:496;

S1V Ln 1:378 0:320:

The cross ratio and distance for the scale point are


mSCL
V


1r 2
0:725;
1r

SSCL 2Ln

1r
0:320:
1r

Then, normalized distance (12.11) is


r1

S1
1:
SSCL

Example 2 The regime has changed on the value r 21 1. It is necessary to dene


its actual regime parameters.
The distance for this regime is
r12 r11 r121 2:

12.2

Regulation Characteristic of Boost Converter

347

Then, subsequent voltage (12.12) is


VL2 156:25 

1:17422 1  1
2:618  1
69:85:
156:25 
22

1
2:618 1
1
1:174

In turn, voltage change (12.16) is given as


VL21

69:85  48:49
24:841:
1  4  0:0064 69:8548:49
252

On the other hand, as the change VL21 is equal to a unit value, the change is
n 1 by (12.7).
Then, by (12.15), we get the subsequent values
21

n2

1:986 1
2:949;
1 0:0064  1:986  1

D2 0:661:

The values VL ; D; n for the further steps are shown in Fig. 12.9.
It is visible that each time the actual voltage change is reduced and cannot reach
the maximum value.
Example 3 Let the regime with the initial value VL1 48:49 be changed to the
value VLN 1 69:85 by small steps consistently. In turn, the consecutive reduction
of the change step reduces undesirable transients.
Let the number of steps be N = 5. It is necessary to nd the values VLi ; ni on the
each step i.
We nd the hyperbolic distance r 61 corresponding to the initial (the zero step)
regime and nal (the fth step) regime. According to Example 2, this distance is as
follows: r 61 1.
For these ve steps, the regime change (as a hyperbolic distance) equals
Dr 0:2.
The value (or length) of the rst step is
r 2 r 1 r 21 1 0:2 1:2:
The rst step is VL2 according to (12.12) and n2 by (12.18) are given as
1:17422:2  1
52:96;
1:17422:2 1
1 1:174 2:2  1
2:183:
n2
0:08 1:1742:2 1

VL2 156:25

348

12

Pulse-Width Modulation Regulators

The voltage change on the rst step according to (12.16) is


VL21

52:963  48:49
5:
1  4  0:0064 52:96348:49
252

The changes on the following steps are also equal to this value.
The transformer ratio change on the rst step according to (12.13) is
n21

2:183  1:986
0:202:
1  0:0064  2:183  1:986

For subsequent steps, the values n; VL are obtained via the recurrent relationships
(12.15), (12.16), since the changes on the following steps keep their values
2:183 0:2029
2:379;
1 0:0064  2:183  0:2029
52:963 5
VL3
57:34;
1 4  0:0064 52:9635
252
n3

n4 2:574;
VL4 61:617;

n5 2:768;
VL5 65:78;

n6 2:96;
VL6 69:84:

It is visible, as such actual changes of the values n,VL are decreased.

12.3

Regulation Characteristic of BuckBoost Converter

12.3.1 BuckBoost Converter with an Idealized Choke


Let us consider a PWM buckboost converter with a given load resistance RL in
Fig. 12.10.

Fig. 12.10 Buckboost


converter with an idealized
choke

VL

V
Ri
L

RL

CL

V0

PWM

Vreg

12.3

Regulation Characteristic of BuckBoost Converter

349

The static regulation characteristic of the idealized converter for the continuous
current of the choke L has the well-known view:
VL V

D
:
1D

If we take into account an internal resistance Ri ; the regulation characteristic is


given by
VL V0

D1  D
2

1  D

Ri
RL

V0

D
1D
 D 2
1 r2 1D

12:21

For convenience, we use the same converter parameters. Then, characteristic


(12.21) has the view of Fig. 12.11.
Similar to (12.2), we introduce the same value
n

1
 1:
1D

Then
VL V0

n1
1 n  12 r2

This expression is close to (12.4). Therefore, we may introduce the value


ncon n  1

Fig. 12.11 Example of


regulation characteristic via
the pulse width

1
D
VL
1
:
1D
1D
V

12:22

160

VL
120

80

40

0
0,0

0,2

0,4

0,6

0,8

1,0

350

12

Pulse-Width Modulation Regulators

This value, at that ncon  0, is the transformation ratio of the idealized converter
itself.
Finally, we obtain
VL V0

ncon
1 r ncon 2

12:23

This equation corresponds to (12.4). Thus, we may use the above results.
Regulation characteristic (12.23) is given in Fig. 12.12. The working area involves
the zero points ncon ,VL .
Next, we may use ellipse Eq. (12.5) at once
r2 VL 2 V2  VV0 0:
The plot of this ellipse is given in Fig. 12.13.
Fig. 12.12 Example of
regulation characteristic via
the converter transformation
ratio

160

VL
80

-80

-160
-20

Fig. 12.13 Geometrical


model of the regulation
characteristic of the buck
boost converter

-15

-10

-5

10

15

ncon

ncon

V0

1 +1

V0
2
0

VL V0

VL

20

ncon
n

12.3

Regulation Characteristic of BuckBoost Converter

351

The values ncon , n turn out at the expense of a stereographic projection too. The
maximum values are
VL M 

V0
;
2r

1
ncon M  :
r

12:24

The regime change or load voltage regulation is dened by a group hyperbolic


transformation, which consecutively, step by step, translates an initial point VL1 to a
subsequent point VL2 and so on. The conformity of the characteristic points and
running points of various variables is shown in Fig. 12.14.
Also, we will use a cross ratio of four points. For that, let us determine necessary
points of characteristic regimes. There are two points of the maximum voltage and a
unit or starting point
VL 0;

ncon 0;

n 1;

D 0:

As it is visible, this case of a unit point differs from above case (12.7).

Fig. 12.14 Regime change as


a group transformation and
conformity of the
characteristic and running
points of the variables
VL ; n; D

V0

V0
2

1+
V0

1+

21

VL

VL

VL

V0

VL

21

n
1

1
1

0.5

1+ 1

1
1+

n
1

D
21

mV
0

SCL

1/mV
S

SCL

SCL

mV

mV

mV

SCL

mV

352

12

Pulse-Width Modulation Regulators

For the initial values n1con , n1 , VL1 , the corresponding cross ratios are similar to
(12.8), (12.9)




1
1
1
1
1
m1n  n1con 0
n1 1 1
r
r
r
r


1
1
1
1
n

1


n
con
 r ;
r1

12:25
1
1
1

n
1

con
r
r n

m1V

 V0
VL1
V0 1
V0
VL 0

m1n 2 :
2r
V0
1
2r
2r

V
L
2r

12:26

Because the values D, n are connected among themselves by the fractionally


linear expression, it is possible to express cross ratio (12.25) via the value D
m1D m1n




1
1
D1 0
r1
r1

1
1
r1 D
1
1
r1  D

r1
:
r1

12:27

Expressions (12.25) and (12.26) lead also to identical values if we take the
logarithm
VL
:

VL
2r

V0

S1 Ln m1V Ln m1n ;

S1 Ln 2r
V0

In particular, for a unit point, this distance is equal to zero.


Besides the considered three points of the characteristic regimes, there is still a
fourth one, the scale point:
ncon 1;

VL

V0
1 r2

n 2;

D 0:5:

This scale point should be considered too. The cross ratio and hyperbolic distance for the scale point are as follows:
mSCL
V
SSCL

!
V0
V0
V0
1 r2

0
;

2
2r 1 r
2r
1  r2
1r
:
Ln mSCL
2Ln
V
1r

12:28

Also, we will note that as such a characteristic fourth point, it is possible to


accept the symmetric point, that is,
ncon 1;

VL

V0
1 r2

n 0;

D 1:

12.3

Regulation Characteristic of BuckBoost Converter

353

This point is also specied in Fig. 12.14. In this case, the cross ratio will be the
inverse value to the above received value, but the hyperbolic distance will be
identical in the absolute value.
Further, it is natural to introduce the normalized hyperbolic distance for the
running regime (the index 1 is lowered), using the obtained scale
r

S
SSCL

VL
1r
:
 2Ln
1r
2r  VL

V0

12:29

Ln 2r
V0

Thus, the normalized distance considers all the characteristic points.


In turn, the inverse expression is
 r2r
1
V0 11r
VL
:


2r 1 r 2r 1

12:30

1r

For a regime change, n1con ! n2con , we have


m21
n



1 2
1
1
 ncon ncon
r
r
1
n2con r1 n1con r

1
 n2con r1  n1con

1
r
1
r

n2con n1con
1r2 n2con n1con
n2con n1con
1r2 n2con n1con

It is possible to introduce, similarly to (12.13), the transformation ratio change


n21

n2con  n1con
1  r2 n2con n1con

n2  1  n1  1
1  r2 n2  1n1  1

Then
n21
:
21
rn

r
m21
n 1

12:31

The subsequent value


n2con

n1con n21
1 r2 n1con n21

12:32

354

12

Pulse-Width Modulation Regulators

Similarly, for the voltage change, VL1 ! VL2 , it is possible to write down at once
VL21
;
21
2r  VL

V0

2r
m21
V V0

VL21

VL2  VL1
1  4r2

VL2 VL1
V0 2

VL2

VL21 VL1
1 4r2

VL21 VL1
V0 2

12:33

Let us consider the change m21


D . Using (12.27), we have
2
1
m21
D mD  mD

1
2
r1 D
1
2
r1  D

1
1
r1 D
1
1
r1  D

12:34

where the change


D21 n21

D2
1D2

D
 1D
1

D
D
1  r2 1D
2 1D1
2

D2  D1
1  D2 D1 D2 D1 1  r2

12:35
:

Thus, the mutually coordinated system of all the regime parameters turns out.
Similar to Fig. 12.6 and relationships (12.18)(12.20), we can use the linearization of the regulation characteristic.
For that, we give, similar to (12.30), the expression for the nonlinear function
calculator as
Vreg D

 1 r r

1
1r

:
1r r
1 r 1r 1  r

12:36

In turn, we rewrite (12.29) for variable as


VL
1r
:
 2Ln
1
r

V
L
2r

V0

Vfb Ln 2r
V0

12:37

12.3.2 BuckBoost Converter with Losses of Choke


Let us consider a buckboost converter with an idealized voltage source V0 in
Fig. 12.15 [7].

12.3

Regulation Characteristic of BuckBoost Converter

355

Fig. 12.15 Buckboost


converter with a loss
resistance of choke

VL
Ri
RL

CL
V0

PWM

Vreg

Taking into account a loss resistance R of choke L, the regulation characteristic


is set by well-known expressions
VL V0

D1  D
2

1  D r

V0

n1
1 rn2

12:38

The plots of these dependences are close to Figs. 12.11 and 12.12.
Similar to (12.22), let us introduce the variable V so that
VL
n  1:
V

12:39

Then, we obtain the general equation of ellipse as


r2 VL2 1 r2 V2 2r2 VL V  V0 V 0:

12:40

The plot of this ellipse is shown in Fig. 12.16. The value n turns out at the
expense of a stereographic projection.
Let us obtain the characteristic values, using the data of the previous example.
So, we give the ready formulas for the maximum values
VLM

V0
V0
p 2
2
2rr  1 r 2r nM

144:2 VLM
25

;

2  0:080:08  1:0032
169:24 VLM
p (
13:54 nM
r 1 r2

nM
:
r
11:54 n
M
We accept the points VL 0; n 1 as a unit points, and VL 25; n 0 as the
scale points.

356

12

Pulse-Width Modulation Regulators

Fig. 12.16 Geometrical


model of the regulation
characteristic of the buck
boost converter with the loss
resistance of choke and
conformity of the points
VL ; n; D

nM

V
n=0

V0

n=1
+

nM

169.24

VLM
V0 25

144.2

VL
mV

0.726
0

VLM

0.852

mn
+

DM

DM

1.086

0.926

Then, the cross ratios for the initial regime are


VL1  VLM
VLM
  ;
1
VL  VLM VLM

1
n  n
M 1  nM
n1 1 nM 1
:


n  nM 1  nM

m1V VLM
VL1 0 VLM

m1n n
M

The cross ratios for the scale points are


V0  VLM
VLM
  0:726;
V0  VLM VLM
n 1  nM
M

0:852:
nM 1  n
M

mSCL

V
mSCL
n

All the above relationships are applicable further.


Also, we may use the linearization of the regulation characteristic. For this, we
give the required expression for the nonlinear function calculator

Vreg


p2r 1
1 r= 1 r2
1
p 1r=p
1 r2
D 1 r2  r 1 r2 
p2r 1
1 r=p
1 r2
1
2
1r= 1 r

12:41

12.3

Regulation Characteristic of BuckBoost Converter

357

and for the inverse nonlinear function calculator


p

Vfb r

V0
 VL r  1 r2
p
Ln 2r
V0
2
2r  VL r 1 r

p
1 2rr 1 r2
p :
 Ln
1 2rr  1 r2

12:42

References
1. Erickson, R.W., Maksimovic, D.: Fundamentals of Power Electronics. Springer, Berlin (2001)
2. Iskender, I., Genc, N.: Design and analysis of a novel zerovoltage-transition interleaved boost
converter for renewable power applications. Int. J. Electron. 97(9), 10511070 (2010)
3. Lo, Y.K., Chen, J.T., Lin, C.Y., Ou, S.Y.: Improved control-to-output characteristics of a
PWM buck-boost converter. Int. J. Circuit Theory Appl. 39(2), 203209 (2011)
4. Lui, X., Wang, P., Loh, P.: A Hybrid AC/ DC micro grid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
5. Michal, V., Premont, C.H., Pillonet, G.: Switched DC/DC boost power stage with linear
control-to-output conversion ratio, based on the ramp-modulated PWM generator. European
patent EP 2482433 A2, 1 Aug 2012
6. Penin, A.: Method of regulating the voltage with step-up and inverting pulse converters. MD
Patent 4067, 26 Aug 2008
7. Penin, A.: Regimes analysis of the voltage pulse regulators on the basic of the invariance
property of the control characteristics. Probl. Reg. Energ. 1 (2009)
8. Penin, A.: Analysis of regimes of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Systems. 12(1), 12 (2013) http://www.wseas.
org/wseas/cms.action?id=6933. Accessed 30 Nov 2014
9. Sira Ramirez, H., Rios-Bolivar, M.: Sliding mode control of dctodc power converters via
extended linearization. IEEE Trans. Circuits Syst.1. Fundam. Theory Appl. 41(10),
652660 (1994)
10. Sundareswaran, K., Devi, V., Nadeem, S.K., Sreedevi, V.T., Palani, S.: Buckboost converter
feedback controller design via evolutionary search. Int. J. Electron. 97(11), 13171327 (2010)

Part IV

Circuits with Non-Linear Load


Characteristics

Chapter 13

Power-Source and Power-Load Elements

13.1

Introduction

In the electric circuit theory, the concepts of power-source and power-load elements
are developed on the basis of generalization of results in power processing systems
[14, 8]. Analysis of a power supply system, which contains a power-load element,
shows the two-valued voltage of a limited capacity voltage source. But for all that,
the volt-ampere characteristic of this power-load element has one-valued representation. On the other hand, taking into account the losses of real power-source
and power-load elements, we get the two-valued volt-ampere characteristic of these
elements [6]. Also, the influence of supply line losses onto the power-load element
has practical importance [5, 7].
Therefore, it is desirable reasonably to determine the regime parameters of the
power-source and power-load element using the single-valued area of their characteristics; that is, to present the regime parameters in the relative form.

13.2

Two-Valued Regime of a Regulated Converter. The


Concept of a Power-Source and Power-Load Element

Let us consider the simplest power supply system with a voltage regulator (stabilizer) VR and given load resistance R1 in Fig. 13.1. The load power
P

V1 2
IV;
R1

or

I P=V

represents a hyperbola in Fig. 13.2.

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_13

361

362

13

Power-Source and Power-Load Elements

I1

VR
V0

Ri

V
V1=
I
R1

Fig. 13.1 Power supply system with a given load power

Fig. 13.2 Power-load


element P and its hyperbolic
characteristic

So, we have a power-load element P for different values of voltage V and current I.
In turn, the voltage source characteristic
I

V0 V
 ;
Ri Ri

13:1

~
as a straight line, intersects the hyperbola into two points M, M.
Using load power characteristic I P=V, we get the quadratic equation
V0 V V2

P:
Ri
Ri

13:2

Therefore, we have the corresponding two roots


V0
V
2

s!
4PRi
1 1 
;
V0 2

~ V0
V
2

s!
4PRi
1 1
:
V0 2

13:3

On the contrary, we may say about a power-source element, which gives a


constant power into a variable load in Fig. 13.3.

13.2

Two-Valued Regime of a Regulated Converter.

363

Fig. 13.3 Power-source


element P and its hyperbolic
characteristic

Note
The offered graphic representation of both power-load P and power-source
P elements corresponds by analogy to a voltage source and source of current.
Further, we will consider power-load elements using the results [5, 6].

13.3

Influence of Voltage Source Parameters


and Power-Load Element onto a Power Supply
Regime

13.3.1 Ideal Voltage Source


The power-load element P with an ideal voltage sourceV0 is represented in
Fig. 13.4. A volt-ampere characteristic of P is the known hyperbola I P=V.
In turn, a volt-ampere characteristic of V0 represents a vertical line with the
coordinate V V0 . A point of intersection I; V0 of this line and hyperbola
determines a single or one-valued regime of this circuit.

Fig. 13.4 Power-load


element and an ideal voltage
source

364

13

Power-Source and Power-Load Elements

For a different voltage source value V01 , a new point I 1 ; V01 will be the energy
equivalent point to the initial point I; V0 because this circuit does not possess the
own or internal scales, which can specify the qualitative characteristics of an
operating regime. It is true for the innite large values I 1; V 1 of our ideal
circuit. Therefore, regimes of different comparable circuits are equivalent always.

13.3.2 Voltage of a Power Supply with Limited Capacity


Let us consider the simplest power supply system in Fig. 13.5. For different voltage
source values (V01 , V02 , and so on), we get the voltage source characteristic as
~ 1 ; M2 ,
parallel lines. These lines intersect the hyperbola into pairs of points M1 , M
~ 2 , and so on. The arrows show the conformity of these points. Also, the charM
acteristic points M , M  correspond to the tangent lines V0M . The closed arrows
~ V0 =2,
illustrate these xed points. For the xed points, from (13.3), when V V
the following condition takes place
1

4PRi
V0 2

0:

Therefore, we get the allowable minimum voltage source values


p
V0M 2 PRi :

Fig. 13.5 Characteristics of a


power-load element with
different values of a voltage
source

13:4

I
Ri

V0

M2

P
M1

M
-VM

M1

0
VM

M2
V

V0M V 1

-V0M

V
2

V0

13.3

Inuence of Voltage Source Parameters and Power-Load Element

365

The corresponding power-load element voltages


p
VM  PRi :

13:5

We must prove the single-valued area of the power-load element characteristic.


To do this, we consider the characteristics of the power-load element in the
projective coordinates in Fig. 13.6. The parallel lines of the voltage source characteristics are intersected into point S onto the innitely remote line 1. This point S
is a pole and the straight line M M  is a polar. Therefore, we get some symmetry
or mapping of the lower part of our curve onto the upper part. The points
M ; M  are the xed or base points. So, the obtained single-valued area involves
the characteristic points VM ; 1; VM .
Now, using the conformity of the voltages V, V0 , we may represent the regime
parameters in the relative form.
From (13.2), we get
V0

Ri P
V:
V

13:6

This dependence determines a hyperbola in Fig. 13.7. Therefore, the


single-valued mapping of the hyperbola points onto the axis V takes place.
Let us constitute the cross-ratio m1V for the initial point V 1

 V 1  VM
m1V VM V 1 1  VM 1
;
V VM

13:7

where the points VM V20M ; VM  V20M are base ones and V 1 is a unit point.
The conformity of the points V 1 , m1V is shown in Fig. 13.8.
Fig. 13.6 Characteristics of
the power-load element in the
projective coordinates

M1

M1

VM

V0M

-V0M
S

-VM

366

13

Power-Source and Power-Load Elements

Fig. 13.7 Dependence V0 V


for a given load power

Fig. 13.8 Conformity of


different regime parameters

V
0

VM

21

-VM

-V0M

V0

mV

21

V0

V0M

V0

V0

21

mV
0

mV

mV

1
2
Further, the cross-ratio m21
V , which corresponds to a regime change V ! V ,
has the form


m21
V

V0M 2 1
V0M
V V 
2
2

V 2  V20M V 1  V20M

:
V 2 V20M V 1 V20M

13:8

Using the normalized values


2
 2 2 V [ 1;
V
V0M

1
 1 2 V [ 1;
V
V0M

we get regime change (13.8) as


2  1 V
1  1

 V
2 1
m21
 1
:
V 1 V V  1 2
V 1 V 1

13:9

13.3

Inuence of Voltage Source Parameters and Power-Load Element

367

This expression we rewrite in the view


2  1 V
 1 1 VV22V1V1
V
1 1

V 2 V 1 1

m21
:
V
2
1


V 1 V  1
 2 V 1  1
V

13:10

Then, we introduce the voltage change V 21 expression


 2V
1  1
V
V 21  2  1 :
V V

13:11

So, the following relationships are performed


m21
V

V 21 1
;
V 21  1

V 21

m21
V 1
:
m21
V 1

13:12

Using (13.11), we obtain the subsequent voltage


21  1
2 V V  1 :
V
1
V 21  V

13:13

 1 1, then the
There is a group transformation. Moreover, if the initial value V
2
21

subsequent value V 1 regardless of the value V .
Similarly to the above, let us consider the cross-ratio m10 for the voltage V0 using
the conformity of the variables by Fig. 13.8 for the single-valued area. This
cross-ratio has the form

 V 1  V0M
m10 V0M V01 1  V0M 01
:
V0 V0M

13:14

Also, the following equality takes place


m10 m1V 2 :

13:15

We may use the hyperbolic metric to determine a regime value or distance


H 1 Ln m10 2Ln m1V :

13:16

The base points V0 ; V0 and points VM ; VM , as the limit points, correspond to
the innitely large distance. Therefore, this limit regime has a clear physical sense.
Similarly to (13.8), the cross-ratio m21
0 , which corresponds to a voltage source
1
2
regime change V0 ! V0 , has the view

 V02  V0M V01  V0M
2
1
m21

:
0 V0M V0 V0  V0M 2
V0 V0M V01 V0M

13:17

368

13

Power-Source and Power-Load Elements

Using the normalized values


2
02 V0 [ 1;
V
V0M

1
01 V0 [ 1;
V
V0M

we get the regime change in the view


02  1 V01  1

 V
2 1
m21

:
0 1 V0 V0  1  2
V0 1 V01 1

13:18

This expression we rewrite in the view

m21
0

2  1 V
1 1
V
02
  01

V0 1 V0  1

 2V
1
V
0 0 1
1
 2 V
V

0
0
 2V
 1 1
V
0 0
 2 V 1
V
0
0

1
1

13:19

Similarly to (13.11), we introduce the analogous expression for the voltage


source change V021
 2V
1  1
V
V021 02 0  1 :
V0  V0

13:20

Then, the following relationships are performed


m21
0

V021 1
;
V021  1

V021

m21
0 1
:
m21
0 1

13:21

Using (13.20), we obtain the subsequent voltage


21  1
02 V0 V0  1 :
V
01
V021  V

13:22

Then, there is a strong reason to introduce a voltage source change as V021 and
a voltage change of power-load element as V 21 .
The validity of such denitions for the changes is conrmed by the following
expression similar to initial expression (13.6); that is,
2V021

1
V 21 :
V 21

Example Let the circuit parameters be given as V0 = 5, Ri = 2, P = 3.

13.3

Inuence of Voltage Source Parameters and Power-Load Element

Power-load voltages (13.3)


(
"
r#
3V
5
432
55  1
1 1

:
V

2
~
2
5
2 5
2V
p
p
p
Voltages (13.5), VM  PRi  3  2  6.
Cross-ratio (13.7) for the initial regime V 1 3
m1V

p
3  6 0:5505
p
0:10102:

3 6 5:4494

Let the subsequent regime be given by V02 8.


Then, subsequent power-load voltages (13.3) equals to V 2 7:1623.
Regime change cross-ratio (13.8)
m21
V

V 2  V20M V 1  V20M

0:4903  0:10102 4:8536:
V 2 V20M V 1 V20M

The normalized values


 2 7:1623
p 2:924;
V
6

 1 p3 1:2247:
V
6

Voltage change (13.11)


V 21

2:924  1:2247  1
1:519:
2:924  1:2247

We check expression (13.12)


m21
V

1:519 1
4:8536:
1:519  1

Let us check subsequent voltage (13.13)


 2 1:519  1:2247  1 0:8604 2:924:
V
1:519  1:2247
0:2942
Now, we consider the cross-ratio for the source voltage V0 .
Initial regime cross-ratio (13.14)
m10

p
5  2 6 0:101
p
0:0102:

9:9
52 6

Equality (13.15), 0:0102 0:101022 .

369

370

13

Power-Source and Power-Load Elements

Distance (13.16)
H 1 Ln 0:0102 2Ln 0:10102 4:5848:
Voltage source regime change (13.17)
m21
0

p
82 6
p  0:0102 0:2404  0:0102 23:5578:
82 6

The normalized values


8
02 p
1:633;
V
2 6

1
5
01 V0 p
1:021:
V
V0M 2 6

Voltage source change (13.20)


V021

1:633  1:021  1 0:6666

1:0887:
1:633  1:021
0:6124

We check expression (13.21)


m21
0

V021 1 1:0887 1
23:5578:

V021  1 1:0887  1

Let us check subsequent voltage (13.22)


21  1
02 V0 V0  1 1:0887  1:021  1 0:1111 1:633:
V
01
1:0887  1:021
0:068
V021  V

13.3.3 Internal Resistance of a Voltage Source


Let us consider the circuit with a variable internal resistance in Fig. 13.9. For its
different values, we get the voltage source characteristics as a bunch of straight
lines. The bunch center S conforms to the voltage V0 .
~ 1 . The arrow
For example, the line R1i intersects the hyperbola into points M1 , M
shows the conformity of these points. Also, the characteristic point M corresponds to the rst tangent lines SM for RiM . For this value RiM , from (13.3), as
~ V0 =2, we get
V V
RiM

V0 2
;
4P

VM

V0
:
2

13:23

13.3

Inuence of Voltage Source Parameters and Power-Load Element

Fig. 13.9 Characteristics of a


power-load element with
different values of an internal
resistance

371

I
Ri

V0

M1

Ri

RiM
M1

M
0

S
1

V V0

VM

We must prove the single-valued area of the power-load element characteristic.


To do this, we consider the characteristics of the power-load element in the projective coordinates in Fig. 13.10.
Using the bunch center, the point S, we may draw the second tangent lines SM  .
The point M  corresponds to the innitely remote point 1. This point S is the pole
and the straight line M M  is the polar.
Therefore, we get mapping of the lower part of our curve onto the upper
part. The points M ; M  are the xed or base points. So, the obtained
single-valued area involves the characteristic points VM ; V0 ; 1.
Now, using the conformity of the values V, Ri , we may represent the regime
parameters in the relative form.
From (13.2), we get
Ri

V0 V V2

:
P
P

13:24

This dependence Ri V determines a parabola in Fig. 13.11.


Fig. 13.10 Characteristics of
the power-load element in the
projective coordinates

S
VM

V0

372

13

Power-Source and Power-Load Elements

Ri
RiM

VM

RiM

Ri

mV

V0

Ri

mV

Ri

VM V 1 V0

Fig. 13.11 Dependence Ri V for a given load power and conformity of different regime
parameters

Therefore, the single-valued mapping of the parabola points onto the axis V
takes place.
Using expression (13.7), we may constitute the cross-ratio m1V for the initial
point V 1 ; that is,

 V 1  VM V 1  V0 =2
:
m1V VM V 1 V0 1

V0 =2
V0  VM

13:25

The points VM V20M ; V 1 are base ones and V0 is a unit point. The
conformity of the points V 1 , m1V is shown in Fig. 13.11.
Similarly to the above, let us consider the cross-ratio m1i for the resistance Ri
using the conformity of the variables by Fig. 13.11 for the single-valued area. This
cross-ratio has the form

 RtiM  R1i
m1i RiM R1i 0 1
:
RiM

13:26

Also, the following equality takes place


m1i m1V 2 :

13:27

13.3.4 Power of a Power-Load Element


Let us consider the circuit with a variable power of power-load element P in
Fig. 13.12.
For its different values, we get the load power characteristic as a bunch of
hyperbolas. The voltage source characteristic is the known straight line.

13.3

Inuence of Voltage Source Parameters and Power-Load Element

Fig. 13.12 Characteristics of


a power-load element with
different values of its power

373

I
V0

Ri

M1

V
P

PM
+

M1
0
V

VM

V0

~ 1 . Also, the
For example, this line intersects the hyperbola P1 into points M1 , M

characteristic point M corresponds to the tangent hyperbola PM . From (13.23),


we get
PM

V0 2
;
4Ri

VM

V0
:
2

13:28

We must prove the single-valued area of the power-load element characteristic.


To do this, we may use expression (13.24). In this case, we get the similar form
P

V0 V V2

:
Ri
Ri

13:29

This dependence PV determines a parabola in Fig. 13.13. Therefore, the


single-valued mapping of the parabola points onto the axis V takes place, which
coincides with Fig. 13.11.
So, we may use the same cross-ratio (13.25) for the initial point V 1 at once

 V 1  V0 =2
:
m1V VM V 1 V0 1
V0 =2

13:30

The conformity of the points V 1 , m1V is shown in Fig. 13.13.

P
PM

VM

PM

mV

V0

mV

VM V 1 V0

Fig. 13.13 Dependence PV for a given voltage source and conformity of different regime
parameters

374

13

Power-Source and Power-Load Elements

Similarly to the above, let us consider the cross-ratio m1P for the power P1 using
the conformity of the variables by Fig. 13.13 for the single-valued area. This
cross-ratio has the form

 PM  P1
:
m1P PM P1 0 1
PM

13:31

Also, the following equality takes place


m1P m1V 2 :

13.4

13:32

Power-Load Element with Losses

13.4.1 Series Loss Resistance


Let the power-load element contain a series loss resistance R0 in Fig. 13.14. We
must obtain equations of the superposed characteristics of the voltage source and
the power-load element. The known power-load characteristic has the view
VI P=I:
Using (13.1), we obtain the following expression for the current
I

V0 V
V0
P



:
R0 R0 R0 IR0

From here,
V0 I

Fig. 13.14 Variable voltage


source, power-load element
with a series loss resistance
and their characteristics

P
IR0 :
I

13:33

V0 V
R0
V0

V
+

V0(I)

V0
1

V0

V(I)

V0M
VM

IM

13.4

Power-Load Element with Losses

375

Evidently, this equation contains the sum of the hyperbola and straight line.
For different voltage source values (V01 , V02 and so on), we get the voltage source
characteristic as parallel horizontal lines. The pairs of the currents I 1 , ~I 1 , and
~ 1 correspond to the given source voltage V01 . Therefore, the
voltages V 1 , V
volt-ampere characteristic V0 I of this voltage source is the two-valued one.
From (13.33), we get
V0 V

PR0
V:
V

Obviously, this expression coincides with (13.6) and corresponds to Fig. 13.7.
Therefore, all expressions (13.7)(13.22) remain in force.

13.4.2 Two-Port Loss Circuit


Let the losses be given by a two-port in Fig. 13.15.
Let us obtain an equation of the volt-ampere characteristicV0 I0 . We rewrite
expression (1.27) of a two-port
"

I0
I1

"

Y00  Y10

# "


Y10  Y11

V0

V1

where Y parameters
Y00 y10 y0 ;

Y11 y10 y1 ;

I1
V1

y10
V0

y0

y1

YL

I0

Y10 y10 :

YIN
I0
YIN

YL

I0
V0(I0)

V0

0
+

V0 V1

V1(I1)
V0
V1

Fig. 13.15 Variable voltage source, power-load element with loss two-port and their
characteristics

376

13

Power-Source and Power-Load Elements

The matrix determinant


DY Y00 Y11  Y10 2 :
Further, we will consider the bilateral two-port with Y00 Y11 . Next, we use the
transmission parameters. In this case, expression (1.30) has the view
2

3

V0
4 I0 5 chc
shy
q
ch2 c

Y00 Y11
Y10

c is the attenuation coefcient, q


Admittance transformation (1.31)
YIN

3
V1
shc 4 5
 I1 ;
chc
q


sh2 c

DY
Y10 2

13:34

p
DY is the characteristic admittance.

Y00 YL DY
;
Y11 YL

YL
YIN
q thc

;
q
1 YqL thc

13:35

where load conductivity YL I1 =V1 .


The inverse expression of (13.34) is
2

3

V1
4 I1 5 chc
shy
q

3
V0
shc 4 5
 I0 :
chc
q


From here, we nd the power of the power-load element


V1

I1 P1
I0
I0

V0 chc  shc  V0 shc chc


q
q
q
q
"
#
2
I0
I0
 V0 2 2 shc  chc V0 chc2 shc2 :
q
q

Finally, we obtain
V0 2 q

I0 2
chc2 shc2
P1
 V0 I0

0:
q
shc  chc
shc  chc

13:36

For a given power P1 , this dependence V0 I0 determines a hyperbola in


Fig. 13.15.

13.4

Power-Load Element with Losses

377

The equations of the asymptotes to this hyperbola has the view


I0
I0
thc; V0
or
q
q  thc
qV0
; I0 qV0  thc:
I0
thc

V0

The pairs of input conductivities YIN , Y~IN , and the pairs of currents I0 , ~I0 correspond to a given source voltage V0 that determines the above two-valued
volt-ampere characteristic V0 I0 . In that sense, the value YIN denes the
single-valued regime.
The maximum efciency value KP and the minimum input power value P0 MIN
corresponds to the admittance matching YL YIN q; that is,
KP chc  shc2 ;

P0 MIN P1 =KP :

13:37

The corresponding voltages of the maximum efciency


V1

s
P1

;
q

V0

s
P0 MIN

:
q

13:38

For different voltage source values, we get the voltage source characteristic as
parallel vertical lines in Fig. 13.16. These lines intersect the hyperbola into pairs of
~ 1 by the voltage V01 . The arrow shows the
points, for example, the points M1 , M
conformity of these points. Also, the characteristic points M , M  correspond to
the tangent lines V0M . The closed arrows illustrate these xed points.
Fig. 13.16 Two-valued
regime of a variable voltage
source

I0
M1

M1

V0M 0

V0

V0 M V0

V0

378

13

Power-Source and Power-Load Elements

To nding the xed points, we obtain the following expression from (13.34)
V0 V1 chc

I1
P1
shc V1 chc
shc:
q
qV1

This expression is similar to (13.6) and Fig. 13.7. We suppose the derivation
equals zero; that is,
dV0
P1
chc 
shc 0:
dV1
qV1 2
From here, we obtain
V1M

s
P1
thc:

q

13:39

Therefore, we get the allowable minimum voltage source values


V0M

s
P1
thc 2V1M chc:
2 chc
q

13:40

We must show the single-valued area of the voltage source characteristic. To do


this, we consider this characteristic in the projective coordinates in Fig. 13.17.
The parallel tangent lines V0M intersect at the point S as the innitely remote
point 1 onto the axis I0 . This point S is the pole and the straight line M M  is the
polar.
Therefore, we get mapping of the lower part of our curve onto the upper
part. The points M ; M  are the xed or base points. So, the obtained
single-valued area involves the characteristic points V0M ; 1; V0M and point V0
too.

Fig. 13.17 Characteristic of


the variable voltage source in
the projective coordinates

I0

S
M

V0M V 0

-V0M

V0

13.4

Power-Load Element with Losses

Fig. 13.18 Conformity of


different regime parameters

379
+

V0M

V0

V1M

V1

m0

m0
+

V0M

V1

V0

V0

V1M V1

m0

Further, we may use all expressions (13.7), (13.8), (13.14)(13.17). The conformity of all the points is shown in Fig. 13.18.
But, we must take into account the characteristic points for KP ; that is, the
voltages V1 , V0 .
Let us consider the cross-ratio mV for the voltage V1 similarly to (13.7)

 V  VM
mV VM V1 1  VM 1
:
V1 VM

13:41

Similarly to (13.14), we consider the cross-ratio m0 for the voltage V0



 V  V0M
m0 V0M V0 1  V0M 0
:
V0 V0M

13:42

We use also hyperbolic metric (13.16)


H Ln m0 2 Ln mV :

13:43

Further, we may introduce the normalized distance or relative deviation from the
matched regime
D1

H1
Ln m10
Ln m1V

H
Ln m0
Ln mV

13:44

In this case, the value H is a scale value. The conformity of D1 is presented in


Fig. 13.13 too.
Example Let the circuit in Fig. 13.15 be given by the following parameters
Y10

p
5 2:2361;

Y00 Y11 y0 y10

p
5:25 2:2913;

P1 8:

380

13

Power-Source and Power-Load Elements

Using (13.34), we get


DY Y00 Y11  Y10 2 5:25  5 0:25;
p
p
chc 1:05 1:0247; shc 0:05 0:2236;
p p
thc 0:2182; q DY 0:25 0:5:
According to (13.37)
KP 1:0247  0:22362 0:6417;
P0 MIN P1 =KP 8=0:6417 12:466:
Maximum efciency voltages (13.38)
V1

s r
P1
8
4;

0:5
q

V0

s r
P0 MIN
12:466

4:9932:

0:5
q

Minimum voltages (13.39) and (13.40)


V1M

s
r
P1
8
0:2182 1:8685;
thc 

0:5
q

V0M 2V1M chc 2  1:8685  1:0247 3:8293:


Let the initial regime point be given as
V11 8;

I11 P1 =V11 1:

Using (13.34), we get


V01 8:6448;

I01 1:9191:

Cross-ratios (13.7) and (13.14)


m1V

V11  V1M 8  1:8685


0:6213;

8 1:8685
V11 V1M

m10

V01  V0M 8:6448  3:8293


0:386 0:62132 :

8:6448 3:8293
V01 V0M

Hyperbolic metric (13.16)


H 1 Ln m10 Ln 0:386 0:9519:

13.4

Power-Load Element with Losses

381

Cross-ratio (13.42) and its hyperbolic metric (13.33)


m0

V0  V0M 4:9932  3:8293


0:1319;

4:9932 3:8293
V0 V0M

H Ln m0 Ln 0:1319 2:0257:
Relative deviation from the matched regime (13.44)
D1

13.5

H 1
0:9519
0:4699:

H 2:0257

Power Supply Line with Losses

Let the power-load element be connected to a voltage source V0 via T network in


Fig. 13.19.
For different resistance r0 values, we get the voltage source characteristic as a
bunch of straight lines in Fig. 13.20. These load straight lines intersect at a point G.
~ 1 . The arrow
For example, the line r01 intersects the hyperbola into points M1 , M
shows the conformity of these points.
The characteristic point M corresponds to the rst tangent line GM for r0 .
Also, the characteristic point M  corresponds to the second tangent line GM 
for r0 . The closed arrows illustrate these xed points.
We must prove the single-valued area of the power-load element characteristic.
To do this, we consider the characteristics of the power-load element in the projective coordinates in Fig. 13.21.
Therefore, the point G is the pole and the straight line M M  is the polar. Then,
we get some symmetry or mapping of the lower part of our curve onto the
upper part. The points M ; M  are the base points.
So, the obtained single-valued area involves the characteristic points


V1M
; 1; V1M
. We must determine all the characteristic values for voltage V1 and
resistance r0 .

r0
V0

r1
r10

Fig. 13.19 Supply line with a variable series resistance

V1
P
+

382

13

Power-Source and Power-Load Elements

Fig. 13.20 Characteristics of


a power-load element with a
variable series resistance of
supply line

I1

r0

M1

r0

r0

Ri =0

M1

V1M

r0

V1 M V 1

V1

VL

r0

R i=

IL

Fig. 13.21 Characteristics of


the power-load element in the
projective coordinates

I1

r0
V

r0
M

r0

V1M

r0

V1

V1M V
1
G

As it was shown above, the load straight lines intersect at the point G, which has
coordinates ILG ; VLG . Physically, it means that regime parameters do not depend on
the value r0 ; that is, the current across this element is equal to zero at the expense of
the load voltage.
Therefore, it is convenience to use the generalized Thvenin/Helmholtz equivalent generator of our network shown in Fig. 13.22. Then, the parameters of this
generator are the following values
V0
;
r10

13:45

r1 r10
V0 :
r10

13:46

ILG
VLG

13.5

Power Supply Line with Losses

383

I1

Ri
G

IL

VL

V1
+

Fig. 13.22 Equivalent circuit for the power-load element with the supply line

The internal resistance


R i r1

r0 r10
:
r0 r10

13:47

In turn,
r0

r10 Ri  r1
:
r10  Ri  r1

13:48

The resistances Ri , r0 have the following characteristic values


RVi 0;

r0V 

RIi 1;

r10  r1
;
r10 r1

r0I r10 :

13:49
13:50

Thus, an equation of straight line, passing through the point G, has the form
I1

VLG  V1
 ILG :
Ri

13:51

From here, we nd the power-load


P V1 I1 V1

VLG  V1
 V1 ILG :
Ri

13:52

VLG  V1
:
P V1 ILG

13:53

Then, we get
Ri V1

This dependence Ri V1 determines a hyperbola in Fig. 13.23.

384

13

Power-Source and Power-Load Elements

Fig. 13.23 Dependence


Ri V1 for a given load power

Ri

Ri

Ri
1

Ri

V1M

V1M

V1

VL

V1

Let us obtain the extreme values Ri ; R


and the corresponding values
i


; V1M
. We suppose that
V1M
dRi VLG  2V1 VLG  V1 V1 ILG


0:
dV1
P V1 ILG
P V1 ILG 2
The solution of this equation gives the two roots

V1M

1

p
1 VLG ILG =P
;
ILG


V1M

1 

p
1 VLG ILG =P
:
ILG

13:54


Substituting (13.54) in (13.53), we get Ri ; R
i . Using (13.48), we obtain r0 ; r0 .
The conformity of the points V1 , Ri , r0 is shown in Fig. 13.24.

Fig. 13.24 Conformity of


different regime parameters

V1M
+

Ri

V1M V1

Ri

Ri

Ri

Ri

r0

r0

r0

r0

mV

r0

mV

VL

r0

V1

H
D

mV
1

H
1

Ri

13.5

Power Supply Line with Losses

385

Let us constitute the cross-ratio m1V for the initial point V11 ; that is,

 1
 V11  V1M

m1V V1M
V1 1 V1M
;
1

V1 V1M

13:55


where the points V1M
; V1M
are base ones and V1 1 is a unit point. The mV
values are shown in Fig. 13.24.
Similarly to the above, let us consider the cross-ratio m1i for the resistance R1i
using the conformity of the variables by Fig. 13.24 for the single-valued area. This
cross-ratio has the form


 R1i  Ri
1
m1i Ri R1i 1 R
:
i
Ri  R
i

13:56

Next, let us consider the cross-ratio m10 for the resistance r01 using the conformity
of the variables by Fig. 13.24. This cross-ratio has the form

 r 1  r0 r0I  r0
m10 r0 r01 r0I r0 01
 I
:
r0  r0
r0  r0

13:57

Also, the following equality takes place


m1i m10 m1V 2 :

13:58

We use hyperbolic metric (13.16)


H 1 Ln m1i Ln m10 2Ln m1V :

13:59

Further, we must support the characteristic points VLG , RVi , r0V . Then

 G
 VLG  V1M

mVV V1M
G
VL 1 V1M
;

VL V1M

 RVi  Ri Ri
V
mVi Ri RVi 1 R
;
i
Ri
Ri  R
i

V
I

 r  r0
r0  r0

:
mV0 r0 r0V r0I r0 0V
r0  r0 r0I  r0

13:60

Hyperbolic metric (13.16)


H V Ln mVi Ln mV0 2Ln mVV :

13:61

386

13

Power-Source and Power-Load Elements

Further, we may introduce the normalized distance


D1

H1
:
HV

13:62

In this case, the value H V is a scale value. The conformity of D1 is presented in


Fig. 13.24 too.
Similarly, we may consider the other variable resistances of this line.
Example Let the circuit in Fig. 13.19 be given by the following parameters
V0 10;

r10 5;

r1 1; P1 3:

Parameters (13.45) and (13.46) of the generalized generator


ILG

10
2;
5

VLG

15
10 12:
5

Characteristic values (13.49) and (13.50)


RVi 0;

r0V 

51
0:8333;
51

RIi 1; r0I 5:

Extreme voltage values (13.54)

V1M
3

1

q
1 122
3
2

3;


V1M
3

1  3
6:
2

Substituting these values in (13.53), we get the corresponding extreme values

Ri V1M

VLG  V1M
12  3
3;
G 3
33  2
P V1M IL

R
i 6

12 6
12:
362

Using (13.48), we obtain


r10 Ri  r1
53  1
3:3333;

r10  Ri  r1 5  3  1
r10 R
512  1
i  r1
9:1666:
r0

r10  R
5
 12  1

r

1
i
r0

Let the initial regime point be given as


P 3;

V11 4;

r01 3:0882:

13.5

Power Supply Line with Losses

387

Using (13.42), we get


R1i r1

r01 r10
2:909:
r01 r10

Cross-ratios (13.55)(13.57)
m1V

V11  V1M
43
0:1;


46
V11 V1M

R1i  Ri
2:909  3
0:01 0:12 ;

1

2:909  12
Ri  Ri
r 1  r0 r0I  r0
m10 01
 I
r0  r0
r0  r0
3:0882  3:3333 5  3:3333

0:02  2 0:01:

3:0882 9:1666 5 9:1666


m1i

Hyperbolic metric (13.59)


H 1 Ln 0:01 2Ln 0:1 4:6052:
For the characteristic points VLG , RVi , r0V , cross-ratios (13.60)

VLG  V1M
12  3
0:5;

G

12 6
VL V1M
R
3
mVi i
0:25 0:52 ;
12
Ri
r V  r0 r0I  r0
 I
mV0 0V
r0  r0
r0  r0
0:8333  3:3333 5  3:3333

0:25:

0:8333 9:1666 5 9:1666

mVV

Hyperbolic metric (13.61)


H V Ln 0:25 2Ln 0:5 1:3863:
Normalized distance (13.62)
D1

4:6052
3:3219:
1:3863

388

13

Power-Source and Power-Load Elements

References
1. Cid-Pastor, A., Martinez-Salamero, L., El Aroudi, A., Giral, R., Calvente, J., Leyva, R.:
Synthesis of loss-free resistors based on sliding-mode control and its applications in power
processing. Contr. Eng. Pract. 21(5), 689699 (2013)
2. Emadi, A., Khaligh, A., Rivetta, C.H., Williamson, G.: Constant power loads and negative
impedance instability in automotive systems: denition, modeling, stability, and control of
power electronic converters and motor drives. IEEE Trans. Veh. Technol. 55(4), 11121125
(2006)
3. Griffo, A., Wang, J., Howe, D.: Large signal stability analysis of DC power systems with
constant power loads. In: Vehicle Power and Propulsion Conference 2008. VPPC08. IEEE,
16 (2008)
4. Maksimovic, D., Jang, Y., Erickson, R.W.: Nonlinear-carrier control for high power factor
boost rectiers. IEEE Trans. Power Electron. 11(4), 578584 (1996)
5. Penin, A.: Analysis of regime changes of constant powerload. Elektrichestvo 12, 4349 (2008)
6. Penin, A.: Constant power sources and its properties. Elektrichestvo 4, 6065 (2010)
7. Sergeev, B., Romash, A., Nagovitsyn, V., Kurchenkova, N.: Analysis of electric circuit with
constant powerload. Elektrichestvo 6, 1622 (2002)
8. Singer, S., Erickson, R.W.: Powersource element and its properties. IEEE Proc. Circ. Devices
Syst. 141(3), 220226 (1994)

Chapter 14

Quasi-resonant Voltage Converter


with Self-limitation of Load Current.
Similarity of Load Characteristics of Some
Electronic Devices

14.1

Load Curve of an Active Two-Pole


with Self-limitation of the Current

In the above Part I, we have considered active two-poles with the load straight lines.
In turn, there are active two-poles with self-limitation of the output current, which
are characterized by nonlinear load curves [13]. So-called the zero-current
switching load resonant converters have these load curves [9]. The typical form
of this load curve is represented by a convex line 1 in Fig. 14.1. In the rst
quadrant, the active two-pole gives energy; the corresponding regime is changed
from the short circuit SC to the open circuit OC. Thus, depending on degree of
convexity, curve 1 can alter from line 2 to line 3.
In the second and fourth quadrants, the active two-pole consumes energy, but
there is limitation of the current even for the high load voltage.
Examples of load quasi-resonant converters
Example 1 It is known the load-resonant converters, for example, with the
zero-current switching [3, 9]. These converters regulate their output by changing of
a switching period TS. Let us consider DC resonant converter with given elements
in Fig. 14.2. It has self-limitation of the output current if the switching period
p
TS 2Tr; the value Tr 2p Lr Cr1 Cr2 is the own oscillation or resonance
period.
Let us suppose ORCAD simulation for TS = 32 S, Tr = 16 S [15]. The load
curve for the rst quadrant is almost rectangular and represents two obvious areas.
Area 1 corresponds to a voltage source; the load voltage depends on its current a
little. Area 2 corresponds to the start of the current limitation and determines the
maximum load power point PM. Areas 3, 4 correspond to a current source. In
particular, point 4 denes SC regime. So, the limitation of the load current takes
place.
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_14

389

390

Quasi-resonant Voltage Converter with Self-Limitation

14
I
I

SC

3
2

3
V

OC

V
1
3

Fig. 14.1 Typical load lines of an active two-poles. 1 Convex. 2 Linear. 3 Rectangular

i T1

(a)

(b)

Cr1

T1

0.4 F

2V0

Cr2

250V

8 H

T2

Lr

30

2
PM

20

0.4 F

I, A
1

10

I
V

V0

0
0

40

RL

(c)

80

80

120

V, Volt

60

40

iT1 , A

1
20
0

t
-20
-40

Tr =16

Fig. 14.2 Known quasi-resonant converter. a Electric circuit. b Load curve with characteristic
areas by numbers 14. c Superposed current pulses of one transistor for these characteristic areas

14.1

Load Curve of an Active Two-Pole

391

The current pulses for these characteristic areas explain the line convex and its
horizontal part. The forward pulse amplitude 1 is determined by a light load, which
changes from OC to the maximum load power. The maximum amplitude corresponds to a point PM and equals about 80 A. Area 3 determines the overload regime
for a heavy load, the forward amplitude is decreased, and the reverse pulse appears.
For SC regime, the reverse amplitude is equal to the forward amplitude 4 if we do
not consider losses. We note that the total amplitude of forward and reverse pulses
does not change for areas 24 and equals about 80 A.
Example 2 The proposed DC resonant converter in Fig. 14.3 with given elements
has the natural self-limitation of the output current [14]. The switching period
p
TS Tr; the value Tr 2p Lr1 Cr 43 lS is the own oscillation period.
Fig. 14.3 Proposed
quasi-resonant converter.
a Electric circuit. b Load
curve with characteristic areas
by numbers 16. c Current
pulses of both transistors for
characteristic areas 3, 4

(a)
iT1

Lr1

Lr2

36 H

T1

36 H

T2

Cr

RL

2V0

1.3 F

150 V

voltage load

iT 2

resistive load

(b)

30
5

2
PM

44

20

I, A

TSTr

TS , S

1
2

50
PM

10
4

5
0

20

(c) 40
20

88
40
V, Volt

PM
60

V0 80
6 6

iT1

iT2

iT1

iT1,iT2, A 0

iT2
t

-20
-40

Tr /2=21.5 S

392

14

Quasi-resonant Voltage Converter with Self-Limitation

Let us consider ORCAD simulation, for the rst quadrant, of the load curves (a
resistive load) with the different switching period TS [15]. The period TS = 44 S is
the minimum working value and Ts = 88 S is the doubled period.
Area 1 for all the curves corresponds to a voltage source too. Area 2 corresponds
to the start of the current limitation and determines the maximum load power point
PM. Area 3 in the form of an inclined line corresponds to the further current
limitation up to SC point 4.
The current pulses for the characteristic areas 3, 4 explain the load line convex
and its inclined part. Area 3 determines the overload regime for a heavy load; the
forward amplitude does not decrease but the short reverse pulse appears.
For SC regime, the reverse pulse extends and its amplitude equals to the forward
current amplitude.
In the general case, for construction of the load line in all the quadrants, it is
necessary to use an alternating voltage source of variable amplitude and corresponding frequency as a load. Therefore, in the second and fourth quadrants, the
converter transfers power from this voltage load to the own voltage source 2V0. In
this case, there is a limitation of the current even for the high load voltage too. It is
clear for areas 5, 6. In particular, for area 5, the reverse pulse amplitude is being
increased more than the forward current amplitude. In turn, the forward current is
absent for area 6 and the converter is simply a rectier.
Further, we note that the load curve for the rst quadrant is almost rectangular
for the doubled period TS = 88 S too. On the contrary, if the period TS Tr, SC
current rises and the load line aspire to a straight line.
The represented load characteristics allow carrying out the analysis and justifying an equivalent generator or circuit of such an active two-pole with
self-limitation of output current.
In addition, similarity of characteristics of such a quasi-resonant converter and
characteristics of different nonlinear elements as transistors, solar cells, and so on
attracts attention.

14.2

Equivalent Generator of an Active Two-Pole


with Self-limitation of Current

Such points of the characteristic regimes as SC and OC are suggesting the possibility of using an equivalent generator with a nonlinear internal resistance Ri1 in
Fig. 14.4 [13]. The voltage source VOC and SC current ISC
1 determines the corresponding value of the internal resistance
RSC
i1

V0OC
:
I1SC

14.2

Equivalent Generator of an Active Two-Pole

393

Then, it is necessary to determine the dependence of the internal resistance on


the voltage. Therefore, it is worth noting three regions of the above represented
characteristics.
For the rst region, when TS 2Tr, after the point PM until SC point and further
in the area V 0, the current is increasing a little. Therefore, the equivalent
generator characteristic is similar to a current source.
The third region, when TS Tr, corresponds to resonance, the characteristic has
the linear form. Therefore, SC current is dened by a small loss resistance. Thus,
the equivalent generator is a voltage source.
For the second region, when Tr < TS 2TS, there is an intermediate case; that is,
the equivalent generator type changes from a voltage source to current source.
Therefore, the characteristic changes from a curve similar to vertical line to a
curve similar to horizontal line.
The rst region of characteristics
Let us consider one curve of the load characteristics. To establish the dependence of
the internal resistance on the voltage, we dene the resistance value Ri1(V) for the
actual simulated data of the current and voltage and build a plot of this relationship
in Fig. 14.5; that is, line 1.
A strictly linear dependence, which denes the minimum resistance value ROC
i1 ,
attracts our attention. This value corresponds to the voltage VOC.
Further, we will dene the load resistance RL(PM) for the maximum load power
PM according to the actual data. It turned out that this value is close to the internal
resistance value (as for the known equivalent circuit) but for SC regime
RL PM RSC
i1 :

14:1

Accepting the linear dependence of line 1, it is possible to obtain an approximation expression of a load characteristic and analytically to check condition
(14.1). It will be the weighty logical justication of the above equivalent generator.
For obviousness, lines 2, 3 show different dependences of Ri1(V) in Fig. 14.5 too.
If the resistance Ri1(V) is not dependent on the voltage (the straight line 2), we get a
voltage source with this internal resistance and a usually linear load characteristic in
Fig. 14.1; that is, line 2. If the resistance Ri1(V) goes through the point VOC
Fig. 14.4 Equivalent
generator and its load
characteristic

XS
Ri2

I2

SC

SC

AI1 +I2
I

SC

S
SC

Ri1
V

OC

I1

I1

RL

SC

I2

I1
I2

AV

OC

I
0

OC

V
YS

394

14

Quasi-resonant Voltage Converter with Self-Limitation


Ri1

SC

Ri1

3
OC

Ri1

OC

AV

OC

Fig. 14.5 Internal resistance dependence of the typical active two-pole. 1 With self-limitation of
the load current. 2 Real voltage source. 3 Ideal current source

(the straight line 3), we have a current source with the rectangular characteristic in
Fig. 14.1; that is, line 3.
Then, the given case represents the intermediate version, when the straight line 1
passes through the point (AVOC, 0). Line 1 equation has the following appearance
for the work area, V VOC,
Ri1



V
OC
V

:
A
I1SC
1

14:2

On the other hand, we obtain from the equivalent circuit that


Ri1

V OC  V
:
I1

14:3

Eliminating value Ri1, we get a load curve equation


I1 V I1SC A

1V
:
AV

14:4

Here and further, the value V is a normalized voltage by the value VOC.
Expression (14.4) denes a hyperbole. The parameter A determines degree of
convexity of curve 1 in Fig. 14.1 for the working area V VOC. If A , this
curve degenerates into a straight line in Fig. 14.1; that is, line 2. If A = 1, we get a
rectangular characteristic; the hyperbola merges with its asymptotes (Fig. 14.1;
lines 3).
The load power has the view
P
1V
;
AV
A
V
PSC
G

14:5

14.2

Equivalent Generator of an Active Two-Pole

395

where PSC
G is the maximum power of the equivalent generator for SC regime. This
expression determines a hyperbola and is similar to efciency expression (4.49).
Therefore, we may use the known results.
The maximum load power
p p2
PM
A A A1 :
SC
PG

14:6

The corresponding load voltage


VPM A 

p
AA  1 VM :

14:7

From here, we may nd the value A.


If A , hyperbola (14.5) transforms to known parabola (2.49). If to express
the load resistance RL = V/I1 via voltage V = V+M by (14.4), condition (14.1) is
conrmed.
Similarly, we may get the load curve equation for the working area V VOC.
The second region of characteristics
Area 3 of characteristics becomes more linear with a greater inclination if TS Tr.
Therefore, the characteristics have the linearly hyperbolic form.
Then, we may make an assumption about the introduction of a linear resistor Ri2
in the equivalent circuit in Fig. 14.4. The resistor value Ri2 is specied by the period
TS. A physical support of this element Ri2 is that, in the third area, its value is so
small that the load current is set by this resistance in a greater extent.
Let us obtain an approximation expression of the load characteristic for this case.
Under the equivalent generator, taking into account (14.3) and (14.4), the load
current is given as the sum of the following hyperbolic and linear components:
IV I1 V I2 V I1SC A

1V
I2SC 1  V;
AV

14:8

where SC current for the linear component


I2SC

V OC
:
Ri2

14:9

Expression (14.8) denes a hyperbola with a center S and asymptotes SXS, SYS in
Fig. 14.4 for ISC
1 = 1. In practice, for a given resultant curve I(V), we must nd the
SC
following settings as ISC
1 , A, and I2 . To do this, the system of three Eq. (14.8) for
1
1
2
three pairs of points (I , V ), (I , V2), and (I3, V3) is being solved.

396

14.3

14

Quasi-resonant Voltage Converter with Self-Limitation

Deviation from the Maximum Load Power Point

The maximum load power corresponds to different load voltage values depending
on the parameter A or proportions between the hyperbolic and linear component of
the currents. For clarity, this voltage coordinate can accept values from (14.7) for
the hyperbolic characteristic up to the value for the straight line one; that is,
1 V+M 0.5. Therefore, likewise to Sect. 4.4, the similar problem arises for scales
and relative expressions of the above deviation.
To do this, we consider regime symmetries of load characteristic (14.8) for a
given value of A in Fig. 14.6. Let the load be changed from SC to OC regime. We
pass to the projective coordinates, which is determined by a center F similarly to
Figs. 2.24 and 4.22. The point F is formed due to the intersection of the tangential
lines or asymptotes FXF, FYF at the xed or base points ISC, VOC. In this case, a
point on the hyperbola is assigned as the rotation of radius vector RFF from the
initial point P+M to a running point or load R1.
Therefore, the pole F and polar TQ (which passes through the points ISC, VOC)
determine the mapping or symmetry of the region of power consumption by the
load on the region of power return; that is, P P.
In turn, the pole T and polar FQ, as a complementary system, determine the
mapping or symmetry of the running points R1, R2 relatively to the maximum power
point P+M. The point T is formed due to the intersection of the tangential line TP+M and
the known line TQ. In this case, the point on the hyperbola is assigned as the rotation
of radius vector RTT from the base point P+M to the running point or load R1.
Further, we will consider the concrete example of the load characteristic in
Fig. 14.3b. For TS = 44 S, the following settings, as ISC
1 = 24.33, A = 1.016, and
ISC
=
9.805
are
obtained.
2
Equation (14.8), for all the normalized currents by ISC
1 = 24.33, has the view
IV 1:016

Fig. 14.6 Symmetry of the


load characteristic relatively
to the maximum power point
P+M

1V
0:4031  V:
1:016  V

14:10

T
SC

A+I2

XS
XF
I

SC

R2

Q
+

PM
RT

R1
RF

SC

I2

I2
0

VQ

1 A
V

OC

YF

V
YS

14.3

Deviation from the Maximum Load Power Point

397

The maximum load power coordinates equal V+M = 0.86, I+M = 0.968. The point
F coordinates have the form
A
0:9845;
2A  1

VF

IF

A
A1
I2SC
0:99:
2A  1
2A  1

To nd the point T coordinates, we use the following equations of the straight


lines TQ and TP+M accordingly:


IV 1 I2SC V 1 I2SC 1:403V 1:403


IV V tga IM  VM tga 1:071V 1:889;

14:11

where the value


tga

dI
A1
A
 I2SC 1:071
dV
A  V2

determines the slope angle for the tangential line TP+M. The solution of these
equations gives VT = 1.464, IT = 3.457.
Similarly, we nd the point Q coordinates using the equations of the straight
lines TQ and FP+M accordingly


IV 1 I2SC V 1 I2SC 1:403V 1:403


IV V tgb IF  VF tgb 0:176V 0:816;

where the value


tgb

IF  IM
0:176
VF  VM

determines the slope angle for the line FP+M. The solution of these equations gives
VQ = 0.371.
Let us map the harmonic conjugate points T, ISC, Q, VOC of the polar TQ onto
the axis V in Fig. 14.7. These points correspond to the points VT, 0, VQ, 1
accordingly.
The mutual mapping of the points VT, VQ relatively to 0, 1 is shown by arrows.
For these harmonic conjugate points, the cross-ratio equals 1; that is,
0 VT VQ 1 0  1:464 0:371 1

1:464  0 0:371  0

1:
1:464  1 0:371  1

Similarly, let us map the harmonic conjugate points Q, P+M, F, PM of the polar
FQ onto the axis V. These points correspond to the points VQ, V+M, VF, VM
accordingly.

398

14

Fig. 14.7 Mutual mapping of


the characteristic points and
its correspondence for the
voltage, cross-ratio, and
distance for the hyperbolic
and linearly hyperbolic cases

Quasi-resonant Voltage Converter with Self-Limitation

VT

hyper.

0.5

VM

VQ

1.464

0.371

0.8

VF

0.86 0.984

V
VM linearly

1.016 1.19

(A1)/A
m(Q) 0

hyper.

0.651

0.182

linearly-hyper.

10.34

hyper.

m(F)

1 m(F)

m(Q)
0.096

-hyper.

D
0

linearly-hyper.

The mutual mapping of the points VF, VQ relatively to V+M, VM is shown by


arrows. The point VM = 1.19 is a point of intersection of the hyperbola second
branch and the line FQ; in Fig. 14.6 it is not shown.
As it is, we get the two conjugate systems of poles and polar lines. Now, it is
necessary to nd the common xed or base points for these conjugate systems.
We note, that the points 0, 1 and V+M, VM are the harmonic conjugate points
shown by arrows. Therefore, we may choose the points 0, 1 as the base ones. Now,
it is necessary to check the cross ratios for each points VF, VQ, using a unit point
V+M; that is,
mQ 0 VQ VM 1 0 0:371 0:86 1 0:096;
mF 0 VF VM 1 0 0:9845 0:86 1 10:34

1
1

:
mQ 0:096

Also, we check the cross ratios for the rest characteristic points VT, V = A
mT 0 VT VM 1 0  1:464 0:86 1 0:096 mQ;
mA 0 A VM 1 0 1:016 0:86 1 10:33 mF 

1
:
mQ

Therefore, we may accept the value m(Q) as a scale value. So, all the characteristic points are evaluated via this scale, as it is shown by the axis m in Fig. 14.7.
So, a running regime point V (for example, V = 0.8) is expressed by the following cross-ratio:
mV 0 V VM 1

V
V
 M
V  1 VM  1

0:8
0:86

4  6:142 0:651:

0:8  1 0:86  1

14:12

14.3

Deviation from the Maximum Load Power Point

399

The corresponding hyperbolic distance from the maximum load power point
HV LnmV 0:428:
Similarly, the scale hyperbolic distance
HQ LnmQ Ln 0:096 2:343 LnmF Ln 10:33:
Then, the normalized distance or relative deviation from the maximum load
power point
D

HV
0:182:
HQ

The inverse expression


mV mQD :
If the value D be given, we nd the actual value V using (14.12) and the known
value m(Q).
Similar expressions are being obtained for the fourth quadrant.
Particular case
If T , we get the hyperbolic load characteristic. Then, the fundamental
expressions obtain the view
r
p
A1
mQ 0
1 0 0:5 A  AA  1 1
;
A
r
r
V
A1
A1
m0 V
;
m mV 0 V VM 1
1V
A
A
VQ VM

14:13

The scale hyperbolic distance


H0:5 Lnm0:5:
The inverse expression
mV m0:5D ; or


V
A  1 D1=2

m0 V
:
1V
A

14:14

If it is necessary to set any equal deviation D for different A, we nd m0(V) using


the second member of (14.14) and then, we get the value V by (14.13). The
corresponding mapping of V, m is shown in Fig. 14.7 too.

400

14.4

Quasi-resonant Voltage Converter with Self-Limitation

14

Symmetrical Load Characteristic for the Full Area


of the Load Voltage Variation

Let us return to Fig. 14.5. Two lines 1 for the corresponding areas V VOC,
V VOC we are replacing by one hyperbola 2 shown in Fig. 14.8a. In this case,
initial Eq. (14.2) obtain the following view


Ri
ROC
i

2


  OC 2
1
Vi 2
r
 2
iOC r 2 ;
OC
A V
Ri

14:15

where Vi = VOC V is the internal resistance voltage and a value rOC


ROC
i
i .
Hereinafter, the subscript 1 for Ri1 and Ii1 is not used.
From here, using the equivalent circuit in Fig. 14.4, we obtain the normalized
load characteristic equation
IVi
Vi =rAV OC
q


2 :
AI SC
A2 r 2 V =rAV OC

14:16

The plot of this expression has the typical view in Fig. 14.8b.
If to introduce the coordinate system ID 0GS VGS, this curve is close to the typical
transfer characteristic ID(VGS) of MOSFET transistors and its approximation by a
hyperbolic tangent [6]. The values ICD, VCGS correspond to the cusp of this curve.
Fig. 14.8 Symmetrical load
characteristic. a Internal
resistance via its voltage.
b Current of the internal
resistance via its voltage and
transfer characteristic.
c Current via internal
conductivity

Ri

(a)
2

Ri
0

(b)
IDM

OC

ID

-V

OC

2V

OC

Yi
OC

yi

V
I
IM
SC

0
V

0DS

Vi

ID

SC

OC

OC

AI
1

OC

ri

(c)

SC

Ri

VGS

OC

VGS

Vi

14.4

Symmetrical Load Characteristic

401

Also, it is possible to note that expression (14.16) is the particular case of Rapps
model of a solid-state microwave power amplier [8, 19].
If to express the current I through the internal conductivity Yi = 1/Ri, yOC
= 1/rOC
i
i ,
the circle equation turns out
 2 
2
Yi
I

1:
14:17
AI SC
yOC
i
The plot of this circle is shown in Fig. 14.8c.

14.5

Asymmetrical Load Characteristics

We may introduce asymmetrical characteristics of a common view too. For


example, a transfer characteristic of MOSFET transistor may considerably be differing from a symmetric curve.
Case 1
We consider a new position of the known hyperbola in Fig. 14.9a.
Let the asymptotes 0y, 0x form a rectangular coordinate system y0x, which turned
on an angle concerning the initial system Ri 0 Vi. Therefore, the equation of the
hyperbola has the view
y k=x:
Fig. 14.9 Asymmetrical load
characteristic. a Internal
resistance via its voltage.
b Current of the internal
resistance via its voltage and
transfer characteristic.
c Current via the internal
conductivity

(a)

14:18
I

(c)

Ri

IM

Ri

Yi

Vi

Vi

YiM

IM

(b)
IDM
C

ID

ID

I
+

IM
I

Vi

0
0DS

IM

VGS

VGS

402

14

Quasi-resonant Voltage Converter with Self-Limitation

On the other hand, for the coordinates of point M, the following orthogonal
transformation is known [10]
" #
x
y

"

cos a

sin k

 sin a

cos a

# "


Vi

Ri

14:19

Then, expression (14.18) obtains the view




cos a  sin a  R2i cos2 a  sin2 a Ri Vi  cos a  sin a  Vi2  k 0:
This view corresponds to the quadratic form [10]
a11 R2i 2a12 Ri Vi a22 Vi2 a33 0:

14:20

Hereinafter, dimensions of the values are not used.


Using transformation of the variable Ri by I, we get a quadratic equation
I2 2

a12 Vi2
a11 Vi2
I
0:
2
a22 Vi a33
a22 Vi2 a33

14:21

The plot of this expression has the typical view in Fig. 14.9b with different
values of the maximum currents I+M, IM.
If to express the current I through the internal conductivity Yi, the general circle
equation turns out
a11 2a12 I a22 I 2 a33 Yi2 0:
From here, we get the explicit circle equation

2 
2
k
Y 2 I  IC 1 IC ;
cos a  sin a i

14:22

where


cos2 a  sin2 a 1 1

 tga :
I
2 cos a  sin a
2 tga
C

The plot of this circle is shown in Fig. 14.9c.


The maximum currents I+M, IM correspond to Yi = 0. Therefore,
IM 1=tga;


IM
tga:

In turn, the current IC conforms to the center of the segment I+MIM and determines
the maximum value YiM.

14.5

Asymmetrical Load Characteristics

403

If asymmetrical curve (14.21) is used for approximation of the transfer characteristic ID(VGS) of MOSFET transistor, it is necessary, in the initial coordinate
system ID 0GS VGS, to restore the coordinate system I 0 Vi. To do this, we may use
the following property of the tangent line into the cusp ICD, VCGS of the asymmetrical
characteristic; this line passes through the origin of the coordinate system I 0 Vi.
This property follows from circle (14.22).
We note, if = 450, the above symmetrical curve is realized. In our case,
expression (14.21) contains two parameters as k, , which realize comprehensive
facilities for approximation.
We get still more facilities, if to use a nonrectangular coordinate system. Then,
similar expression (14.21) will contain three parameters.
Case 2
We may consider the known hyperbola Ri(Vi) in Fig. 14.10a and rewrite Eq. (14.18)
for the coordinate system y0x
y k=x:

14:23

On the other hand, for the coordinates of point M, the following transformation
takes place:
" #
x
y

" #
Vi
0
:

1
Ri

1= cos a

tga

Then, expression (14.23) obtains the view


Ri Vi  tga

k  cos a
:
Vi

14:24

Using transformation of the variable Ri by Yi and I, we get the two equations


Yi

Vi
;
Vi2  tga k  cos a

Vi2
:
Vi2  tga k  cos a

The maximum values


YiM

1
p ;
2 k  sin a

ViM

s
k  cos a
1

tga
2YiM  tga

From here, we get


tga

1
;
2YiM Vi

k  cos a

ViM
:
2YiM

14:25

404

14

Quasi-resonant Voltage Converter with Self-Limitation

Fig. 14.10 Asymmetrical


load characteristic. a Internal
resistance via its voltage. b
Current and internal
conductivity via its voltage.
c Current via internal
conductivity and its
stereographic projection onto
the voltage axis

(b)

(a)

I
1

Ri

0.5

Ri
y

Vi

Yi

x
0

0.5

Vi

Vi

Vi

(c)

I
1

0.5

Vi
0

0.5

Vi

Vi
Yi

Then, expressions (14.25) obtain the relative form


Yi
2 Vi =ViM

;
YiM Vi =ViM 2 1

I
Vi =ViM 2

;
IM Vi =ViM 2 1

where IM = 2I(YiM) = 2YiM ViM. The plots of these curves are shown in Fig. 14.10b
for the normalized values Yi, Vi, I.
If to express the current I through the internal conductivity Yi, the explicit circle
equation turns out


Yi
2YiM

2

I
1

IM 2

2

1
:
4

The plot of this circle is shown in Fig. 14.10c for the normalized values. In turn,
~i are resulted by the known stereographic projection.
the voltages Vi ; V
The considered dependences are similar to the regulation characteristic of the
above voltage regulators in Chap.10 that gives a certain physical meaning to the
parameters of the approximation curves.
Also, these dependences coincide with Salehs models of a modulation characteristic for traveling-wave tube [11, 21].

14.5

Asymmetrical Load Characteristics

405

So, we have obtained a whole class of curves and convenient approximation


expressions for characteristics of different electronic devices.

14.6

Linearly Hyperbolic Approximation of a Solar Cell


Characteristic

14.6.1 Approximation Problem


For the analysis and calculation of power supply systems based on solar cell
generators, a mathematical model of solar cell is needed [20]. The traditional
exponent model requires iterative numerical calculation methods. Therefore, it
might be valuable to use simpler models that allow carrying out the direct or
analytical computations [2, 4, 5, 7].
Other directions use formal approximation. In particular, the quadratic fractional
expression of the model allows calculating the maximum load power point by a
quadratic equation [1]. But, a resistive load leads to a cubic equation that complicates the calculations in real time of working regime; that is, the load current and
voltage.
More convenient a linearly hyperbolic model is similar to Eq. (14.8). The calculation with a resistance load leads to a quadratic equation [16].
Also, this model makes it possible to soundly determine deviation from the
maximum load power point and to compare the effectiveness of different solar cells.
The fact is that in photovoltaic power systems the maximum power point tracker
MPPT is used extensively, for example [12]. Usually, solar array is a set of solar
cell modules connected in parallel and in series. Then, it is possible to use either a
common or individual MPPT for each module.
In the case of the common MPPT, the differences in the module parameters will
appear. This leads to the deviation of the module regime within a large range from
its maximum load power point, although, as a whole, the array works under these
conditions.
If the individual MPPT is used, then this device can be turned on only in the case
of sufciently great regime deviations.

14.6.2 Formal Linearly Hyperbolic Approximation


Let us consider the calculated load curves in Fig. 14.11 according [1].
Numbers 13 denote the typical areas similar to Fig. 14.3b. Area 1 for all the
curves, as inclined lines, corresponds to a voltage source with a relatively high
internal resistance. Area 2 corresponds to the start of the current limitation and

406

14

Fig. 14.11 Example of the


typical solar cell load curves
with different values of
insulation and characteristic
areas shown by number

Quasi-resonant Voltage Converter with Self-Limitation


200

1
150

V, Volt 100

50

0.5
0

0.75 1 SUN
8

10

12

14

I, A

determines the maximum load power point PM. Then, area 3 corresponds to a
current source with nearly horizontal straight line.
Hence, the inverse similarity follows with the curves in Fig. 14.3b; that is, the
inclined area 1 in Fig. 14.11 corresponds to area 3 in Fig. 14.3b. Then, using
Fig. 14.4, it is possible to draw up a dual equivalent generator in Fig. 14.12.
The current generator ISC and the internal nonlinear conductivity Yi1 dene a
hyperbolic component, and the linear conductance Yi2 denes a linear component of
the load voltage. We may write the equation immediately, using expression (14.8)
and changing formally currents I by voltages V and resistances by conductivities
VI V1 I V2 I V1OC A

1  I=I SC
V2OC 1  I=I SC ;
A  I=I SC

14:26

SC OC
where OC voltage for the hyperbolic component is VOC
1 = I /Yi1 , andfor the linear
OC
SC
component is V2 = I /Yi2.
Similarly to Fig. 14.5, we may propose the nonlinear conductivity Yi1 dependence on the load current I in Fig. 14.13 for the dual active two-pole.

Fig. 14.12 Equivalent


generator of a solar cell and
its load characteristic

XS
V

Yi1
I

V1

SC

V
RL

Yi2

OC

V2

OC

AV 1 +V2
OC

S
OC

V1

OC

V2

V1

V2

AI
I

SC

SC

YS

14.6

Linearly Hyperbolic Approximation

407

Yi1

OC

Yi1

3
SC

Yi1

SC

AI

SC

Fig. 14.13 Internal conductivity dependence on the load current of the typical active two-port. 1
With self-limitation of the load voltage. 2 Real current source. 3 Ideal voltage source
OC
For the given curves in Fig. 14.11, it is need to nd the settings A, VOC
1 , and V2
using SC, OC points and two points a, b points and two points a, b, which are on
the each side near to the maximum load power point (a working area) with Ia, Va
and Ib, Vb coordinates.
Let us give the necessary design parameters

aI a
;
ab
Vb

Ib
;
Ia  Ib

 V OC
Va
Vb
; b12

;
b12
1  Ia 1  Ib
1
b12

; V2OC V OC  V1OC :
1
1
A Aia  Ai
b

b 1I
V1OC

The currents Ia, Ib are normalized by the current ISC.


Proposed expression (14.26) allows the direct calculation of the current I at the
given load resistance RL. In this case, the load voltage V = RLI.
Substituting this expression in (14.26), we get the requirement equation


OC
OC
OC
AV OC
SC V1 V2
1 V2
0:
I 2  AI SC
I

AI
RL V2OC =I SC
RL V2OC =I SC

14:27

The maximum power point parameters are found from the cubic equation
2I 3  1 3A AuI 2 2A1 AuI  A2 u 0;
OC
where u = 1 + VOC
1 /V2 .

408

14

Quasi-resonant Voltage Converter with Self-Limitation

Example At rst, we use the traditional characteristic of a solar cell [1]


V RS I

mkT Iph  I I0
Ln
;
q
I0

14:28

where RS is the series resistance of the cell, the values m, k, T, q are the diode factor,
Boltzmann constant, temperature, electron charge correspondingly; the values Iph,
I0 are the photocurrent and reverse saturation current.
For the given values, this expression takes the view
V 0:9I

1
13:615  I 0:0081
Ln
:
0:042
0:0081

14:29

From here, VOC = 176, ISC = 13.608.


In turn, the maximum load power parameters have the following values
PM IM VM 11:32  123:6 1400;

RLM VM =IM 10:92:

Next, we consider proposed expression (14.26).


OC
Then OC voltage, VOC
= 176.
1 + V2
a
a
b
Let I = 12, V = 114.76, I = 10, Vb = 135.58.
Then,
10
114:76
135:58
5; b12

459:8;
12  10
1  0:881 1  0:734
135:58
 176
0:881  5
0:7293; A
1:0324;
b 10:734
459:8
5  0:7293
1
459:81
V1OC

134:7
1
1
1:0324 1:03240:881
 1:03240:7293
a

V2OC V OC  V1OC 41:3:


Finally, we get the approximation expression


I
1  13:608
I
VI 139:06
41:3 1 
:
I
13:608
1:0324  13:608

14:30

The curves V(I) under expressions (14.29), (14.30) are shown in Fig. 14.14.
Examination of these curves shows that proposed model simulates a solar cell
characteristic well for practical tasks. The found errors of the working area (for
example, from Ib to Ia) are less than 2 %.
In turn, the curves of the power P = IV are shown in Fig. 14.15.
Then, if R = 10.92, we get from (14.27)

14.6

Linearly Hyperbolic Approximation

Fig. 14.14 Comparison


between the proposed
(dash-dot line) and traditional
(solid line) model of
volt-ampere characteristics:
a Full scale. b Working area
is near to the maximum power
point

409

(a)

200

150
b

I P
M

V, Volt 100

50

10

12

14

I, A

(b)

140

130

V, Volt

PM

120

110

100

Fig. 14.15 Comparison


between the proposed
(dash-dot line) and traditional
(solid line) model of power
volt characteristics

10

11
I, A

12

13

1500
1250
1000

P, W 750
500
250
0

10

12

14

I, A

I 2  26:97I 177:2 0:
The solution gives IM = 11.33. Using (14.30), we are nding VM = 123.36.
Therefore, this power is equal to 1398 and coincides with the maximum load
power.

410

14

Quasi-resonant Voltage Converter with Self-Limitation

About of Validation of the Linearly Hyperbolic Approximation


As it was shown above, the model parameters are calculated for the concrete curve
(with the appropriate level of insolation) from two points near the maximum load
power point. But, the calculated values depend on coordinates of these points that
complicate the nding the parameters on photocurrent levels. Therefore, there is a
problem of a physical validation of linearly hyperbolic approximation.
An empirical approximation is presented in [18]. This approximation is based on
a possible hyperbolic dependence of pn junction current for its large value.
Deviation from Maximum Load Power Point
The considered results make it possible to soundly determine deviation from the
maximum load power regime and to compare the effectiveness of different solar
cells [17]. To do this, we may use the corresponding expressions of Sect. 14.3.

References
1. Akbaba, M., Alattavi, M.: A new model for I-V characteristic of solar cell generators and its
applications. Solar Energy Mater. Solar Cells 37, 123132 (1995)
2. Alghuwainem, S.: A close-form solution for the maximum-power operating point of a solar
cell array. Solar Energy Mater. Solar Cells 46(3), 249257 (1997)
3. Ang, S., Oliva, A.: Power-switching converters, 2nd ed., Taylor & Francis Group (2005)
4. Cubas, J., Pindado, S., Victoria, M.: On the analytical approach for modeling photovoltaic
systems behavior. J. Power Sources 247, 467474 (2014)
5. Das, A.: Analytical derivation of explicit J-V model of a solar cell from physics based implicit
model. Sol. Energy 86(1), 2630 (2012)
6. Diakov, V.P., Maximchuk, A.A., Remnev, A.M., Smerdov, V.Y.: Entsiklopedia ustroistv na
polevyh transistorah (Encyclopedia of equipments on eld transistors). Solon-R, Moskva
(2002)
7. El Tayyan, A.: An approach to extract the parameters of solar cells from their illuminated I-V
curves using the Lambert W function. Turkish J. Phys. 39(1), 115 (2015)
8. Eltholth, A., Mekhail, A., Elshirbini, A., Dessouki, M., Abdelfattah, A.: A new- trend modelbased to solve the peak power problems in OFDM systems. Model. Simul. Eng. 2008, 5
(2008)
9. Hui, S.R., Chung, H.: Resonant and soft-switching converters. In: Power Electronics Hand
book, pp. 405449. Elsevier Inc. (2006)
10. Korn, G.A., Korn, T.M.: Mathematical Handbook for Scientists and Engineers. MacGraw
Hill, New York (1968)
11. Lee, B.M., de Figueiredo, R.J.: Adaptive predistorters for linearization of high-power am
pliers in OFDM wireless communications. Circ. Syst. Signal Proc. 25(1), 5980 (2006)
12. Ma, J., Man, K., Zhang, N., Guan, S., Wong, P., Lim, E., Lei, C.: Improving
power-conversion efciency via a hybrid MPPT approach for photovoltaic systems.
Elektronika ir Elektrotechnika 19(7), 5760 (2013)
13. Penin, A.: The voltage-current characteristics of an active two-pole with self-limitation of
current. Elektrichestvo 7, 5460 (2008)
14. Penin, A., Semionov, A.: Method of regulated resonance DC-DC voltage conversion.
Europ. patent EP 1 504 517 B1. 05 Nov 2008
15. Penin, A.: A quasi-resonance voltage converter with improved parameters. Elektrichestvo 2,
5864 (2009)

References

411

16. Penin, A., Sidorenko, A.: A convenient model for IV characteristic of a solar cell generator as
an active two-pole with self-limitation of current. World Acad. Sci. Eng. Technol. 3(4), 905
909 (2009)
17. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010)
18. Penin, A.: An empirical validation of the linear-hyperbolic approximation of the IV
characteristic of a solar cell generator. World Acad. Sci. Eng. Technol. 5(3), 13631368
(2011)
19. Rapp, C.: Effects of HPA-Nonlinearity on a 4-DPSK/OFDM-signal for a digital sound
broadcasting system. In: In ESA, Second European Conference on Satellite Communications
(ECSC-2) pp. 179184 (SEE N92-15210 06-32), vol. 1, pp. 179-184 (1991)
20. Rekioua, D., Matagne, E.: Optimization of Photovoltaic Power System. Modelization,
Simulation and Control. Spriger, London (2012)
21. Saleh, A.: Frequency-independent and frequency-dependent nonlinear models of TWT
ampliers. IEEE Trans. Comm. 29(11), 17151720 (1981)

Conclusions

In this book, the features of an electric circuit with changeable operating regimes
are considered. The reader will agree that for such a circuit it is important to
determine the parameters of the running regime in the normalized form using the
characteristic parameter values as scales, to receive the equivalent of this circuit.
But there are two problems.
On the one hand, the actual regime parameters of the load resistance, current,
voltage result in various values of the corresponding normalized quantities. The
hands-on experience as if agrees with that.
On the other hand, interference of loads or any resistance on the load regime
leads to change of scales.
The offered approach to interpretation of regime changes as projective transformation allowed connecting regime parameters in one system and to consider
these changes through an invariant value in the form of the cross ratio of four
points. The reader will agree that the adequate mathematical model of regime
behavior turned out and we obtain the basis for research of such circuits and
introduction of necessary concepts. Naturally, the presented results are only the
beginning of researches. In particular, it is possible to apply such approach to
alternating current circuits.
If we look more widely, it is possible to speak about representation of flowed
processes of the different physical nature, using known electromechanical analogy.
Also, the presented approach is being applied for a long time in other scientic
domains, as mechanics, biology, and so on; for example, it is possible to see the
following publications:
1. Vaseashta, A.K., Penin, A., Sidorenko, A.: On the analogy of non-euclidean
geometry of human body with electrical networks. Int. J. Electr. Comput. Eng.
4(3). 378388 (2014)
2. Stakhov, A.P.: The mathematics of harmony: From Euclid to contemporary
mathematics and computer science. Series on knots and everything, vol. 22,
World Scientic Publishing Co. Pte. Ltd. (2009)

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4

413

Index

A
Active, 276, 279, 284
multi-port, 193, 282
three-port, 210, 282
two-pole, 3, 9, 1113, 17, 29, 30, 34, 35,
42, 56, 60, 70, 82, 83, 86, 88, 98, 184,
190, 276278, 389, 390, 392, 394, 406
two-port, 9, 14, 87, 88, 167, 168, 175, 207,
224, 244, 275, 276, 279, 281, 407
Additive property, 41, 262, 263
Afne transformation, 29, 30, 32, 9799, 102,
241
Approximation, 20, 393, 395, 400, 403405,
408, 410
Asymmetrical load characteristic, 401, 404
Attenuation coefcient, 18, 103, 376
Auxiliary load, 55, 70, 87
B
Balanced network, 190, 196, 263
Base values, 911, 14, 39, 61, 62, 75, 178, 186,
191, 194, 195, 216
Boost converter, 24, 337, 339, 343, 348,
354356
Buck converter, 23
Bunch center, 38, 55, 57, 66, 71, 79, 8486, 89,
174, 176, 182, 188, 197, 199, 200, 254,
256, 277, 281, 285, 370, 371
C
Cascaded connection, 9, 10, 99, 100, 109, 110
Changeable resistance, 9, 55, 58
Characteristic
admittance, 18, 103, 376
regimes, 1, 4, 9, 12, 20, 21, 32, 39, 53, 168,
177, 181, 188, 199, 238, 292, 340, 342,
351, 352, 392

values, 1, 2, 6, 9, 10, 20, 42, 45, 5961, 67,


73, 80, 87, 90, 107, 182, 250, 268, 345,
355, 381, 383, 386
Comparison
of regimes, 148, 149, 157, 158, 199, 224
Conformal plane, 290, 292, 326, 327
Cross ratio, 39, 40, 48, 49, 53, 6165, 68,
7578, 81, 90, 91, 104108, 110112,
114, 116118, 120, 122, 124, 129, 133,
134, 138, 146, 148150, 155157, 159,
175, 178, 186, 216, 260, 296298,
301304, 307, 309, 311, 318, 319, 321,
330, 341, 342, 352, 365, 367, 369, 373,
379, 385, 398
Current source, 58, 59, 72, 73, 86, 87, 281,
285, 389, 393, 394, 406, 407
D
Deviation, 19, 47, 52, 120, 122, 123, 125, 126,
146, 149152, 155, 157162, 379, 381,
396, 399, 405, 410
Distances, 151, 152, 161, 179, 182, 186, 189,
197, 203, 228, 247, 265, 267, 271, 272
Distributed power supply systems, 1, 289
E
Effectiveness, 1, 3, 94, 126128, 130, 131,
133, 135, 137, 195, 405, 410
Efciency, 1, 3, 5, 6, 17, 19, 47, 108, 120, 121,
127, 130, 131, 135138, 377, 380, 395
Energy indices, 131
Equalizing resistor, 141, 147, 153, 160
F
Fixed points, 41, 50, 52, 54, 108, 145, 146,
152154, 161, 162, 216, 293, 295, 296,
364, 377, 381
Four-port circuit, 237, 241, 244

Springer International Publishing Switzerland 2016


A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4

415

416
G
Generalized equivalent circuit, 57, 71, 84, 279
Group property, 40, 63, 69, 76, 109, 297
H
Half
plane, 328, 329, 331, 332
rounds, 328, 330, 331
Harmonic conjugacy, 48, 52, 128
Homogeneous coordinates, 38, 174, 179, 182,
186, 189, 197, 201, 204, 214, 221, 245,
246, 253, 265, 267, 272
Hyperbolic geometry model, 295, 296, 331
I
Increments, 3, 1517, 22, 63, 76
Innitely remote, 38, 48, 50, 133, 136, 174,
203, 248, 257, 295, 328, 365, 371, 378
Influence of loads, 4, 279
Initial regime, 2, 9, 11, 3234, 36, 40, 41, 42,
45, 61, 63, 6769, 75, 76, 80, 81, 90,
98, 101, 107, 114, 169, 172, 196, 197,
200, 232, 240, 243, 255, 271, 296298,
307, 321, 333, 345, 356, 369, 380, 386
InputOutput Conformity, 97, 102, 237, 244,
263
Intermediate change, 8, 16, 207
Internal resistance, 3, 4, 6, 10, 11, 21, 29, 35,
42, 44, 55, 56, 58, 59, 67, 84, 141,
173175, 289, 290, 313, 349, 370, 383,
392394, 400, 401, 404, 405
Invariant, 3032, 34, 35, 37, 39, 55, 6163, 76,
99, 105, 106, 108, 109, 112, 116, 142,
150, 159, 241, 247, 253, 296, 315, 340
L
Limited capacity voltage
source, 21, 22, 141, 337, 361
Linearization, 337, 343, 344, 354, 356
Linearly hyperbolic, 398, 405, 410
Linear stabilization, 11, 228
Load
conductivities, 14, 113, 173175, 182, 190,
209, 244, 250, 251, 268, 273, 275, 282
power, 1, 3, 5, 6, 21, 39, 47, 48, 51, 102,
121, 131135, 289, 290, 361, 362, 372,
389, 392397, 399, 405410
resistances, 4, 1517, 313, 323
straight lines, 9, 11, 14, 4143, 45, 5760,
66, 71, 73, 74, 79, 8385, 89, 98100,
102, 103, 109, 110, 168, 173175, 224,
238, 245, 276, 277, 381, 382, 389
Loss resistance, 23, 91, 338, 355, 374, 393

Index
M
Maximum efciency, 1, 108, 120, 122, 124,
128, 377, 380
Maximum load power, 1, 3, 6, 21, 39, 48, 51,
102, 389, 392, 393, 395, 396, 397, 399,
405, 407, 409, 410
Measured input current, 118, 260
Minimum load resistance, 315, 320, 327
Modular connection, 126, 127, 128, 130
Moving of point, 294, 295, 299, 301, 302, 304,
309311
N
Negative resistance, 91, 92, 315, 316
Nonlinear
load characteristics, 389, 392
regulation curve, 337
Non-uniform coordinates, 174, 178, 181, 183,
186, 188, 207, 210, 212, 221, 245, 246,
250252, 260, 268, 273
Normalized values, 5, 1114, 32, 55, 131,
149151, 157, 158, 297, 299, 305, 317,
319, 321, 328, 366, 368370, 404
Norton equivalent circuit, 44, 70, 71, 78, 79,
84, 88, 89
O
One-sheeted hyperboloid, 324
One-valued, 21, 361, 363
Open circuit, 2, 3, 29, 31, 32, 39, 55, 56, 70,
83, 97, 141, 147, 168, 169, 172, 177,
186, 213, 238, 239, 256, 324, 389
P
Parallel connection, 127, 128
Paralleling, 4, 141143, 153
Parametric stabilization, 91
Passive multi-port, 237
Polar, 50, 52, 53, 121, 123, 124, 133, 136, 365,
371, 378, 381, 396398
Pole, 50, 52, 53, 121123, 133, 136
Power-load elements, 17, 21, 361, 363
Power-source elements, 17, 21, 361363
Power supply, 1, 2, 21, 22, 55, 91, 92, 94, 116,
131, 141, 183, 184, 195, 198, 289, 290,
296, 313, 314, 324, 325, 337, 339, 361,
362, 364, 381, 405
Power transfer ratio, 5, 17, 19, 120
Projection center, 32, 61
Projective
plane, 173, 174, 221
transformation, 29, 37, 38, 39, 41, 61, 102,
104, 105, 109, 142, 145, 149, 154, 155,

Index

417
157, 196, 244, 246, 247, 253, 294, 296,
307, 316

Q
Quasi-resonant, 17, 20, 389392
R
Recalculation formula, 5, 6, 40, 44, 46, 99,
101, 196, 259, 260
Reference triangle, 174, 177, 179, 200, 214,
245247, 253, 264
Regime change, 2, 3, 711, 23, 24, 35, 37,
3943, 45, 46, 62, 63, 65, 68, 69, 76,
78, 81, 82, 90, 91, 99, 100, 102, 107,
115, 252, 259, 296, 297, 299302, 307,
317319, 321, 330, 331, 333, 334,
340342, 346, 347, 351, 353, 366369
Regime symmetry, 47, 121, 123
Regulated converter, 361
Regulation, 22, 290
characteristic, 21, 2325, 289, 296, 297,
337340, 343, 349, 350, 354, 356, 404
of load voltages, 289
Relative form, 1, 3, 4, 7, 11, 19, 23, 25, 39, 47,
146, 155, 175, 180, 181, 188, 201, 202,
278, 361, 365, 371, 404
Running regime, 4, 5, 29, 31, 35, 39, 52, 53,
62, 98, 122, 124, 135, 138, 146, 149,
157, 168, 170, 173, 174, 178, 182, 186,
189, 196, 201, 238, 244, 246, 253, 264,
281, 296, 342, 353, 398
S
Scales, 2, 46, 9, 14, 20, 29, 32, 44, 47, 55, 63,
76, 173, 182, 188, 278, 364, 396
Self-limitation, 389, 391, 392, 394, 407
Short circuit, 2, 3, 29, 31, 32, 39, 55, 56, 70,
83, 88, 97, 142, 167, 168, 177, 185,
186, 213, 238, 254, 256, 276, 279, 281,
283, 389
Six-port, 193, 263, 264
Solar Cell, 17, 20, 289, 392, 405, 406, 408, 410
Space, 41
Stabilization, 11, 13, 9193, 198, 199, 313, 322
Static characteristic, 1
Stereographic projection, 290, 291, 313, 314,
326, 340, 351, 355, 404

Subsequent regime, 9, 32, 36, 62, 64, 67, 68,


75, 77, 80, 81, 90, 91, 99, 101, 115,
210, 218, 230, 251, 259, 297299, 302,
307, 330, 331, 333, 369
Symmetrical load characteristic, 400
Symmetry, 4850, 121124, 133, 136, 365,
381, 396
T
Thvenin equivalent circuit, 19, 42, 56, 57, 65,
66, 84
Transfer of signal, 116, 260
Transformation ratio, 1, 21, 22, 289, 291, 298,
302, 306, 307, 310, 313, 315, 318,
320323, 325, 326, 328, 329, 331, 333,
334, 339, 342, 343, 350, 353
Two-ports, 9, 10, 17, 19, 47, 97, 99, 100, 102,
109111, 121, 127, 130, 227
Two-valued characteristic, 21, 337, 361, 377
U
Unit point, 39, 52, 53, 63, 76, 106, 107, 121,
122, 124, 149, 158, 177, 178, 186, 200,
201, 203, 245, 246, 253, 264, 265, 316,
319, 342, 351, 352, 355, 365, 372, 385,
398
V
Variable voltage source, 29, 97, 99
Voltage, 13
converter, 1, 17, 20, 21, 289, 337, 338
ratio, 19
stabilization, 11, 13, 313
Voltage source, 4, 5, 1113, 21, 29, 30, 38, 43,
47, 48, 56, 58, 59, 72, 73, 86, 87,
9193, 98, 118, 121, 141144, 142,
167, 173175, 183, 192, 193, 195, 197,
200, 147, 153, 237, 238, 262, 263, 281,
289, 315, 316, 324, 354, 362365, 367,
368, 370, 372, 374, 375, 377, 378, 381,
392, 393, 405, 407
Volt-ampere characteristic, 3, 17, 20, 21, 29,
42, 361, 363, 375, 377, 409
Z
Zero-order generator, 59, 67, 73, 80